Showing 1–50 of 92 results for author: Amin, M A

Trajectory Data Mining and Trip Travel Time Prediction on Specific Roads

Authors: Muhammad Awais Amin, Jawad-Ur-Rehman Chughtai, Waqar Ahmad, Waqas Haider Bangyal, Irfan Ul Haq

Abstract: Predicting a trip's travel time is essential for route planning and navigation applications. The majority of research is based on international data that does not apply to Pakistan's road conditions. We designed a complete pipeline for mining trajectories from sensors data. On this data, we employed state-of-the-art approaches, including a shallow artificial neural network, a deep multi-layered pe… ▽ More Predicting a trip's travel time is essential for route planning and navigation applications. The majority of research is based on international data that does not apply to Pakistan's road conditions. We designed a complete pipeline for mining trajectories from sensors data. On this data, we employed state-of-the-art approaches, including a shallow artificial neural network, a deep multi-layered perceptron, and a long-short-term memory, to explore the issue of travel time prediction on frequent routes. The experimental results demonstrate an average prediction error ranging from 30 seconds to 1.2 minutes on trips lasting 10 minutes to 60 minutes on six most frequent routes in regions of Islamabad, Pakistan. △ Less

Submitted 9 July, 2024; originally announced July 2024.

Comments: N/A

Balancing Patient Privacy and Health Data Security: The Role of Compliance in Protected Health Information (PHI) Sharing

Authors: Md Al Amin, Hemanth Tummala, Rushabh Shah, Indrajit Ray

Abstract: Protected Health Information (PHI) sharing significantly enhances patient care quality and coordination, contributing to more accurate diagnoses, efficient treatment plans, and a comprehensive understanding of patient history. Compliance with strict privacy and security policies, such as those required by laws like HIPAA, is critical to protect PHI. Blockchain technology, which offers a decentrali… ▽ More Protected Health Information (PHI) sharing significantly enhances patient care quality and coordination, contributing to more accurate diagnoses, efficient treatment plans, and a comprehensive understanding of patient history. Compliance with strict privacy and security policies, such as those required by laws like HIPAA, is critical to protect PHI. Blockchain technology, which offers a decentralized and tamper-evident ledger system, hold promise in policy compliance. This system ensures the authenticity and integrity of PHI while facilitating patient consent management. In this work, we propose a blockchain technology that integrates smart contracts to partially automate consent-related processes and ensuring that PHI access and sharing follow patient preferences and legal requirements. △ Less

Submitted 2 July, 2024; originally announced July 2024.

Comments: The 21st International Conference on Security and Cryptography (SECRYPT 2024)

BD-SAT: High-resolution Land Use Land Cover Dataset & Benchmark Results for Developing Division: Dhaka, BD

Authors: Ovi Paul, Abu Bakar Siddik Nayem, Anis Sarker, Amin Ahsan Ali, M Ashraful Amin, AKM Mahbubur Rahman

Abstract: Land Use Land Cover (LULC) analysis on satellite images using deep learning-based methods is significantly helpful in understanding the geography, socio-economic conditions, poverty levels, and urban sprawl in developing countries. Recent works involve segmentation with LULC classes such as farmland, built-up areas, forests, meadows, water bodies, etc. Training deep learning methods on satellite i… ▽ More Land Use Land Cover (LULC) analysis on satellite images using deep learning-based methods is significantly helpful in understanding the geography, socio-economic conditions, poverty levels, and urban sprawl in developing countries. Recent works involve segmentation with LULC classes such as farmland, built-up areas, forests, meadows, water bodies, etc. Training deep learning methods on satellite images requires large sets of images annotated with LULC classes. However, annotated data for developing countries are scarce due to a lack of funding, absence of dedicated residential/industrial/economic zones, a large population, and diverse building materials. BD-SAT provides a high-resolution dataset that includes pixel-by-pixel LULC annotations for Dhaka metropolitan city and surrounding rural/urban areas. Using a strict and standardized procedure, the ground truth is created using Bing satellite imagery with a ground spatial distance of 2.22 meters per pixel. A three-stage, well-defined annotation process has been followed with support from GIS experts to ensure the reliability of the annotations. We performed several experiments to establish benchmark results. The results show that the annotated BD-SAT is sufficient to train large deep learning models with adequate accuracy for five major LULC classes: forest, farmland, built-up areas, water bodies, and meadows. △ Less

Submitted 9 June, 2024; originally announced June 2024.

Comments: 26 pages, 15 figures and 12 tables

MCDFN: Supply Chain Demand Forecasting via an Explainable Multi-Channel Data Fusion Network Model

Authors: Md Abrar Jahin, Asef Shahriar, Md Al Amin

Abstract: Accurate demand forecasting is crucial for optimizing supply chain management. Traditional methods often fail to capture complex patterns from seasonal variability and special events. Despite advancements in deep learning, interpretable forecasting models remain a challenge. To address this, we introduce the Multi-Channel Data Fusion Network (MCDFN), a hybrid architecture that integrates Convoluti… ▽ More Accurate demand forecasting is crucial for optimizing supply chain management. Traditional methods often fail to capture complex patterns from seasonal variability and special events. Despite advancements in deep learning, interpretable forecasting models remain a challenge. To address this, we introduce the Multi-Channel Data Fusion Network (MCDFN), a hybrid architecture that integrates Convolutional Neural Networks (CNN), Long Short-Term Memory networks (LSTM), and Gated Recurrent Units (GRU) to enhance predictive performance by extracting spatial and temporal features from time series data. Our rigorous benchmarking demonstrates that MCDFN outperforms seven other deep-learning models, achieving superior metrics: MSE (23.5738%), RMSE (4.8553%), MAE (3.9991%), and MAPE (20.1575%). Additionally, MCDFN's predictions were statistically indistinguishable from actual values, confirmed by a paired t-test with a 5% p-value and a 10-fold cross-validated statistical paired t-test. We apply explainable AI techniques like ShapTime and Permutation Feature Importance to enhance interpretability. This research advances demand forecasting methodologies and offers practical guidelines for integrating MCDFN into supply chain systems, highlighting future research directions for scalability and user-friendly deployment. △ Less

Submitted 23 June, 2024; v1 submitted 24 May, 2024; originally announced May 2024.

Vector Wave Dark Matter and Terrestrial Quantum Sensors

Authors: Dorian W. P. Amaral, Mudit Jain, Mustafa A. Amin, Christopher Tunnell

Abstract: (Ultra)light spin-$1$ particles -- dark photons -- can constitute all of dark matter (DM) and have beyond Standard Model couplings. This can lead to a coherent, oscillatory signature in terrestrial detectors that depends on the coupling strength. We provide a signal analysis and statistical framework for inferring the properties of such DM by taking into account (i) the stochastic and (ii) the vec… ▽ More (Ultra)light spin-$1$ particles -- dark photons -- can constitute all of dark matter (DM) and have beyond Standard Model couplings. This can lead to a coherent, oscillatory signature in terrestrial detectors that depends on the coupling strength. We provide a signal analysis and statistical framework for inferring the properties of such DM by taking into account (i) the stochastic and (ii) the vector nature of the underlying field, along with (iii) the effects due to the Earth's rotation. Owing to equipartition, on time scales shorter than the coherence time the DM field vector typically traces out a fixed ellipse. Taking this ellipse and the rotation of the Earth into account, we highlight a distinctive three-peak signal in Fourier space that can be used to constrain DM coupling strengths. Accounting for all three peaks, we derive latitude-independent constraints on such DM couplings, unlike those stemming from single-peak studies. We apply our framework to the search for ultralight $B - L$ DM using optomechanical sensors, demonstrating the ability to delve into previously unprobed regions of this DM candidate's parameter space. △ Less

Submitted 26 June, 2024; v1 submitted 4 March, 2024; originally announced March 2024.

Comments: 23 pages, 9 figures, and 5 appendices. Compared to v2, added 1 figure. Also, for a movie of the vector field behavior based on simulations, visit https://www.youtube.com/watch?v=bbw6yFRLS7s

Report number: KCL-PH-TH/2024-09

Journal ref: JCAP 06 (2024) 050

Engineering and Revealing Dirac Strings in Spinor Condensates

Authors: Gui-Sheng Xu, Mudit Jain, Xiang-Fa Zhou, Guang-Can Guo, Mustafa A. Amin, Han Pu, Zheng-Wei Zhou

Abstract: Artificial monopoles have been engineered in various systems, yet there has been no systematic study of the singular vector potentials associated with the monopole field. We show that the Dirac string, the line singularity of the vector potential, can be engineered, manipulated, and made manifest in a spinor atomic condensate. We elucidate the connection among spin, orbital degrees of freedom, and… ▽ More Artificial monopoles have been engineered in various systems, yet there has been no systematic study of the singular vector potentials associated with the monopole field. We show that the Dirac string, the line singularity of the vector potential, can be engineered, manipulated, and made manifest in a spinor atomic condensate. We elucidate the connection among spin, orbital degrees of freedom, and the artificial gauge, and show that there exists a mapping between the vortex filament and the Dirac string. We also devise a proposal where preparing initial spin states with relevant symmetries can result in different vortex patterns, revealing an underlying correspondence between the internal spin states and the spherical vortex structures. Such a mapping also leads to a new way of constructing spherical Landau levels, and monopole harmonics. Our observation provides insights into the behavior of quantum matter possessing internal symmetries in curved spaces. △ Less

Submitted 9 April, 2024; v1 submitted 22 February, 2024; originally announced February 2024.

Comments: 5 pages + 5 figures + 5 appendices. In comparison with the previous version, we have (1) added supplementary material as appendices, and (2) made some minor revisions in the main text. Also, a simulation video for creating two artificial Dirac Strings with a mixed spin-1 state, is available at https://youtu.be/PCQGMqd-DfQ

Healthcare Policy Compliance: A Blockchain Smart Contract-Based Approach

Authors: Md Al Amin, Hemanth Tummala, Seshamalini Mohan, Indrajit Ray

Abstract: This paper addresses the critical challenge of ensuring healthcare policy compliance in the context of Electronic Health Records (EHRs). Despite stringent regulations like HIPAA, significant gaps in policy compliance often remain undetected until a data breach occurs. To bridge this gap, we propose a novel blockchain-powered, smart contract-based access control model. This model is specifically de… ▽ More This paper addresses the critical challenge of ensuring healthcare policy compliance in the context of Electronic Health Records (EHRs). Despite stringent regulations like HIPAA, significant gaps in policy compliance often remain undetected until a data breach occurs. To bridge this gap, we propose a novel blockchain-powered, smart contract-based access control model. This model is specifically designed to enforce patient-provider agreements (PPAs) and other relevant policies, thereby ensuring both policy compliance and provenance. Our approach integrates components of informed consent into PPAs, employing blockchain smart contracts to automate and secure policy enforcement. The authorization module utilizes these contracts to make informed access decisions, recording all actions in a transparent, immutable blockchain ledger. This system not only ensures that policies are rigorously applied but also maintains a verifiable record of all actions taken, thus facilitating an easy audit and proving compliance. We implement this model in a private Ethereum blockchain setup, focusing on maintaining the integrity and lineage of policies and ensuring that audit trails are accurately and securely recorded. The Proof of Compliance (PoC) consensus mechanism enables decentralized, independent auditor nodes to verify compliance status based on the audit trails recorded. Experimental evaluation demonstrates the effectiveness of the proposed model in a simulated healthcare environment. The results show that our approach not only strengthens policy compliance and provenance but also enhances the transparency and accountability of the entire process. In summary, this paper presents a comprehensive, blockchain-based solution to a longstanding problem in healthcare data management, offering a robust framework for ensuring policy compliance and provenance through smart contracts and blockchain technology. △ Less

Submitted 15 December, 2023; originally announced December 2023.

General Relativistic Polarized Proca Stars

Authors: Zipeng Wang, Thomas Helfer, Mustafa A. Amin

Abstract: Massive vector fields can form spatially localized, non-relativistic, stationary field configurations supported by gravitational interactions. The ground state configurations (p-solitons/vector solitons/dark photon stars/polarized Proca stars) have a time-dependent vector field pointing in the same spatial direction throughout the configuration at any instant of time, can carry macroscopic amounts… ▽ More Massive vector fields can form spatially localized, non-relativistic, stationary field configurations supported by gravitational interactions. The ground state configurations (p-solitons/vector solitons/dark photon stars/polarized Proca stars) have a time-dependent vector field pointing in the same spatial direction throughout the configuration at any instant of time, can carry macroscopic amounts of spin angular momentum, and are spherically symmetric and monotonic in the energy density. In this paper, we include general relativistic effects, and numerically investigate the stability of compact polarized Proca stars (linear and circularly polarized) and compare them to hedgehog-like field configurations (with radially pointing field directions). Starting with approximate field profiles of such stars, we evolve the system numerically using 3+1 dimensional numerical simulations in general relativity. We find that these initial conditions lead to stable configurations. However, at sufficiently large initial compactness, they can collapse to black holes. We find that the initial compactness that leads to black hole formation is higher for circularly polarized stars (which carry macroscopic spin angular momentum), compared to linearly polarized ones, which in turn is higher than that for hedgehog configurations. △ Less

Submitted 8 September, 2023; originally announced September 2023.

Comments: 10 pages, 6 figures

Morphological Classification of Radio Galaxies using Semi-Supervised Group Equivariant CNNs

Authors: Mir Sazzat Hossain, Sugandha Roy, K. M. B. Asad, Arshad Momen, Amin Ahsan Ali, M Ashraful Amin, A. K. M. Mahbubur Rahman

Abstract: Out of the estimated few trillion galaxies, only around a million have been detected through radio frequencies, and only a tiny fraction, approximately a thousand, have been manually classified. We have addressed this disparity between labeled and unlabeled images of radio galaxies by employing a semi-supervised learning approach to classify them into the known Fanaroff-Riley Type I (FRI) and Type… ▽ More Out of the estimated few trillion galaxies, only around a million have been detected through radio frequencies, and only a tiny fraction, approximately a thousand, have been manually classified. We have addressed this disparity between labeled and unlabeled images of radio galaxies by employing a semi-supervised learning approach to classify them into the known Fanaroff-Riley Type I (FRI) and Type II (FRII) categories. A Group Equivariant Convolutional Neural Network (G-CNN) was used as an encoder of the state-of-the-art self-supervised methods SimCLR (A Simple Framework for Contrastive Learning of Visual Representations) and BYOL (Bootstrap Your Own Latent). The G-CNN preserves the equivariance for the Euclidean Group E(2), enabling it to effectively learn the representation of globally oriented feature maps. After representation learning, we trained a fully-connected classifier and fine-tuned the trained encoder with labeled data. Our findings demonstrate that our semi-supervised approach outperforms existing state-of-the-art methods across several metrics, including cluster quality, convergence rate, accuracy, precision, recall, and the F1-score. Moreover, statistical significance testing via a t-test revealed that our method surpasses the performance of a fully supervised G-CNN. This study emphasizes the importance of semi-supervised learning in radio galaxy classification, where labeled data are still scarce, but the prospects for discovery are immense. △ Less

Submitted 31 May, 2023; originally announced June 2023.

Comments: 9 pages, 6 figures, accepted in INNS Deep Learning Innovations and Applications (INNS DLIA 2023) workshop, IJCNN 2023, to be published in Procedia Computer Science

Journal ref: Procedia Computer Science, Volume 222, 2023, Pages 601-612

i-SPin 2: An integrator for general spin-s Gross-Pitaevskii systems

Authors: Mudit Jain, Mustafa A. Amin, Han Pu

Abstract: We provide an algorithm for evolving general spin-$s$ Gross-Pitaevskii / non-linear Schrödinger systems carrying a variety of interactions, where the $2s+1$ components of the `spinor' field represent the different spin-multiplicity states. We consider many nonrelativistic interactions up to quartic order in the Schrödinger field (both short and long-range, and spin-dependent and spin-independent i… ▽ More We provide an algorithm for evolving general spin-$s$ Gross-Pitaevskii / non-linear Schrödinger systems carrying a variety of interactions, where the $2s+1$ components of the `spinor' field represent the different spin-multiplicity states. We consider many nonrelativistic interactions up to quartic order in the Schrödinger field (both short and long-range, and spin-dependent and spin-independent interactions), including explicit spin-orbit couplings. The algorithm allows for spatially varying external and/or self-generated vector potentials that couple to the spin density of the field. Our work can be used for scenarios ranging from laboratory systems such as spinor Bose-Einstein condensates (BECs), to cosmological/astrophysical systems such as self-interacting bosonic dark matter. As examples, we provide results for two different setups of spin-$1$ BECs that employ a varying magnetic field and spin-orbit coupling, respectively, and also collisions of spin-$1$ solitons in dark matter. Our symplectic algorithm is second-order accurate in time, and is extensible to the known higher-order accurate methods. △ Less

Submitted 2 May, 2023; originally announced May 2023.

Comments: 13 pages, 3 figures, 2 appendices

Journal ref: Phys. Rev. E 108, 055305, 15 November 2023

A novel integrated Sachs-Wolfe effect from Early Dark Energy

Authors: Tristan L. Smith, John T. Giblin Jr., Mustafa A. Amin, Mary Gerhardinger, Ericka Florio, Matthew Cerep, Shar Daniels

Abstract: We study the nonlinear effects of minimally coupled, massless, cosmological scalar fields on the cosmic microwave background (CMB). These fields can exhibit post-recombination parametric resonance and subsequent nonlinear evolution leading to novel contributions to the gravitational potential. We compute the resulting contributions to the CMB temperature anisotropies through the time-variation of… ▽ More We study the nonlinear effects of minimally coupled, massless, cosmological scalar fields on the cosmic microwave background (CMB). These fields can exhibit post-recombination parametric resonance and subsequent nonlinear evolution leading to novel contributions to the gravitational potential. We compute the resulting contributions to the CMB temperature anisotropies through the time-variation of the gravitational potential (i.e., the integrated Sachs-Wolfe (ISW) effect). We find that fields that constitute 5% of the total energy density and become dynamical at $z_c \simeq 10^{4}$ can produce marginally observable ISW signals at multipoles $\ell \simeq 2000$. Fields that become dynamical at earlier times and/or have initial displacements at a flatter part of their potential, produce ISW contributions that are significantly larger and at higher multipoles. We calculate these dynamics and the resulting evolution of gravitational perturbations using analytic estimates alongside detailed nonlinear lattice simulations, which couple scalar fields and cosmological fluids to a perturbed metric. Finally, we discuss the possibility of detecting these features with future high-resolution CMB observations. △ Less

Submitted 18 April, 2023; v1 submitted 4 April, 2023; originally announced April 2023.

Comments: 12 pages, 6 figures, comments welcome; v2 minor typo corrected

Kinetic relaxation and Bose-star formation in multicomponent dark matter- I

Authors: Mudit Jain, Mustafa A. Amin, Jonathan Thomas, Wisha Wanichwecharungruang

Abstract: Using wave kinetics, we estimate the emergence time-scale of gravitating Bose-Einstein condensates/Bose stars in the kinetic regime for a general multicomponent Schrödinger-Poisson (SP) system. We identify some effects of the diffusion and friction pieces in the wave-kinetic Boltzmann equation (at leading order in perturbation theory) and provide estimates for the kinetic nucleation rate of conden… ▽ More Using wave kinetics, we estimate the emergence time-scale of gravitating Bose-Einstein condensates/Bose stars in the kinetic regime for a general multicomponent Schrödinger-Poisson (SP) system. We identify some effects of the diffusion and friction pieces in the wave-kinetic Boltzmann equation (at leading order in perturbation theory) and provide estimates for the kinetic nucleation rate of condensates. We test our analysis using full $3+1$ dimensional simulations of multicomponent SP system. With an eye towards applications to multicomponent dark matter, we investigate two general cases in detail. First is a massive spin-$s$ field with $N=2s+1$ components (scalar $s=0$, vector $s=1$ and tensor $s=2$). We find that for a democratic population of different components, the condensation time-scale is $τ_{(s)}\approx τ_0\times N$, where $τ_0$ is the condensation time scale for the scalar case. Second is the case of two scalars with different boson masses. In this case, we map-out how the condensation time depends on the ratios of their average mass densities and boson masses, revealing competition and assistance between components, and a guide towards which component condenses first. For instance, with $m_1 < m_2$ and not too disparate mass densities, we verify that the time scale of condensation of the first species quickly becomes independent of $m_2/m_1$, whereas for equal average number densities, the emergence time scale decreases with increasing $m_2/m_1$. △ Less

Submitted 19 May, 2023; v1 submitted 4 April, 2023; originally announced April 2023.

Comments: 8 pages + 3 appendices, 5 figures. Videos from simulations are available at https://mustafa-amin.com/home/multicomponent-dark-matter. In comparison with the previous version, we have (1) added more references; (2) provide clarifications in Appendix A and account for an additional factor of 2 in the wave kinetic equation

Journal ref: Phys. Rev. D 108, 043535, 28 August 2023

Automatic Detection of Natural Disaster Effect on Paddy Field from Satellite Images using Deep Learning Techniques

Authors: Tahmid Alavi Ishmam, Amin Ahsan Ali, Md Ahsraful Amin, A K M Mahbubur Rahman

Abstract: This paper aims to detect rice field damage from natural disasters in Bangladesh using high-resolution satellite imagery. The authors developed ground truth data for rice field damage from the field level. At first, NDVI differences before and after the disaster are calculated to identify possible crop loss. The areas equal to and above the 0.33 threshold are marked as crop loss areas as significa… ▽ More This paper aims to detect rice field damage from natural disasters in Bangladesh using high-resolution satellite imagery. The authors developed ground truth data for rice field damage from the field level. At first, NDVI differences before and after the disaster are calculated to identify possible crop loss. The areas equal to and above the 0.33 threshold are marked as crop loss areas as significant changes are observed. The authors also verified crop loss areas by collecting data from local farmers. Later, different bands of satellite data (Red, Green, Blue) and (False Color Infrared) are useful to detect crop loss area. We used the NDVI different images as ground truth to train the DeepLabV3plus model. With RGB, we got IoU 0.41 and with FCI, we got IoU 0.51. As FCI uses NIR, Red, Blue bands and NDVI is normalized difference between NIR and Red bands, so greater FCI's IoU score than RGB is expected. But RGB does not perform very badly here. So, where other bands are not available, RGB can use to understand crop loss areas to some extent. The ground truth developed in this paper can be used for segmentation models with very high resolution RGB only images such as Bing, Google etc. △ Less

Submitted 2 April, 2023; originally announced April 2023.

Comments: 6 pages, 13 figures. This paper has been accepted for presentation at the ICCRE2023 conference, held at Nagaoka University of Technology, Japan

Photons from dark photon solitons via parametric resonance

Authors: Mustafa A. Amin, Andrew J. Long, Enrico D. Schiappacasse

Abstract: Wave-like dark matter made of spin-1 particles (dark photons) is expected to form ground state clumps called "vector solitons", which can have different polarizations. In this work, we consider the interaction of dark photons with photons, expressed as dimension-6 operators, and study the electromagnetic radiation that arises from an isolated vector soliton due to parametric resonant amplification… ▽ More Wave-like dark matter made of spin-1 particles (dark photons) is expected to form ground state clumps called "vector solitons", which can have different polarizations. In this work, we consider the interaction of dark photons with photons, expressed as dimension-6 operators, and study the electromagnetic radiation that arises from an isolated vector soliton due to parametric resonant amplification of the ambient electromagnetic field. We characterize the directional dependence and polarization of the outgoing radiation, which depends on the operator as well as the polarization state of the underlying vector soliton. We discuss the implications of this radiation for the stability of solitons and as a possible channel for detecting mergers of vector solitons through astrophysical observations. △ Less

Submitted 14 May, 2023; v1 submitted 26 January, 2023; originally announced January 2023.

Comments: 30 pages, 2 figures. V2: Updated towards version published in JCAP

Journal ref: JCAP 05 (2023) 015

Cosmological Structure Formation and Soliton Phase Transition in Fuzzy Dark Matter with Axion Self-Interactions

Authors: Philip Mocz, Anastasia Fialkov, Mark Vogelsberger, Michael Boylan-Kolchin, Pierre-Henri Chavanis, Mustafa A. Amin, Sownak Bose, Tibor Dome, Lars Hernquist, Lachlan Lancaster, Matthew Notis, Connor Painter, Victor H. Robles, Jesus Zavala

Abstract: We investigate cosmological structure formation in Fuzzy Dark Matter (FDM) with an attractive self-interaction (SI) with numerical simulations. Such a SI would arise if the FDM boson were an ultra-light axion, which has a strong CP symmetry-breaking scale (decay constant). Although weak, the attractive SI may be strong enough to counteract the quantum 'pressure' and alter structure formation. We f… ▽ More We investigate cosmological structure formation in Fuzzy Dark Matter (FDM) with an attractive self-interaction (SI) with numerical simulations. Such a SI would arise if the FDM boson were an ultra-light axion, which has a strong CP symmetry-breaking scale (decay constant). Although weak, the attractive SI may be strong enough to counteract the quantum 'pressure' and alter structure formation. We find in our simulations that the SI can enhance small-scale structure formation, and soliton cores above a critical mass undergo a phase transition, transforming from dilute to dense solitons. △ Less

Submitted 24 January, 2023; originally announced January 2023.

Comments: 10 pages, 3 figures, submitted to mnras

A lower bound on dark matter mass

Authors: Mustafa A. Amin, Mehrdad Mirbabayi

Abstract: We argue that there is a lower bound of order $10^{-19}$ eV on dark matter mass if it is produced after inflation via a process with finite correlation length. We rely on non-detection of free-streaming suppression and white-noise enhancement of density perturbations as the observational inputs. We argue that there is a lower bound of order $10^{-19}$ eV on dark matter mass if it is produced after inflation via a process with finite correlation length. We rely on non-detection of free-streaming suppression and white-noise enhancement of density perturbations as the observational inputs. △ Less

Submitted 25 February, 2024; v1 submitted 17 November, 2022; originally announced November 2022.

Comments: v2: 4 pages, 2 figures + appendices. Compared to v1: bound made more conservative (by including a logarithmic correction), added explicit examples, updated figures and appendices

i-SPin: An integrator for multicomponent Schrödinger-Poisson systems with self-interactions

Authors: Mudit Jain, Mustafa A. Amin

Abstract: We provide an algorithm and a publicly available code to numerically evolve multicomponent Schrödinger-Poisson (SP) systems with a SO($n$) symmetry, including attractive or repulsive self-interactions in addition to gravity. Focusing on the case where the SP system represents the non-relativistic limit of a massive vector field, non-gravitational self-interactions (in particular spin-spin interact… ▽ More We provide an algorithm and a publicly available code to numerically evolve multicomponent Schrödinger-Poisson (SP) systems with a SO($n$) symmetry, including attractive or repulsive self-interactions in addition to gravity. Focusing on the case where the SP system represents the non-relativistic limit of a massive vector field, non-gravitational self-interactions (in particular spin-spin interactions) introduce complexities related to mass and spin conservation which are not present in purely gravitational systems. We address them with an analytical solution for the `kick' step in the algorithm, where we are able to decouple the multicomponent system completely. Equipped with this analytical solution, the full field evolution is second order accurate, preserves spin and mass to machine precision, and is reversible. Our algorithm allows for an expanding universe relevant for cosmology, and the inclusion of external potentials relevant for laboratory settings. △ Less

Submitted 19 November, 2022; v1 submitted 15 November, 2022; originally announced November 2022.

Comments: 18 pages, 3 figures, 4 appendices. A python code based on our algorithm in provided at, https://github.com/mudit-jain90/i-SPin . Animations of the numerical simulation results can be found at, https://www.youtube.com/watch?v=KUbVtULiBtE . (Added discussions of another algorithm, along with its generalization.)

Journal ref: JCAP 04 (2023) 053, 20 April 2023

Searching for axion-like particles through CMB birefringence from string-wall networks

Authors: Mudit Jain, Ray Hagimoto, Andrew J. Long, Mustafa A. Amin

Abstract: Axion-like particles (ALPs) can form a network of cosmic strings and domain walls that survives after recombination and leads to anisotropic birefringence of the cosmic microwave background (CMB). In addition to studying cosmic strings, we clarify and emphasize how the formation of ALP-field domain walls impacts the cosmic birefringence signal; these observations provide a unique way of probing AL… ▽ More Axion-like particles (ALPs) can form a network of cosmic strings and domain walls that survives after recombination and leads to anisotropic birefringence of the cosmic microwave background (CMB). In addition to studying cosmic strings, we clarify and emphasize how the formation of ALP-field domain walls impacts the cosmic birefringence signal; these observations provide a unique way of probing ALPs with masses in the range $3H_0 \lesssim m_a \lesssim 3H_{\rm cmb}$. Using measurements of CMB birefringence from several telescopes, we find no evidence for axion-defect-induced anisotropic birefringence of the CMB. We extract constraints on the model parameters that include the ALP mass $m_a$, ALP-photon coupling $\mathcal{A} \propto g_{aγγ} f_a$, the domain wall number $N_{\rm dw}$, and parameters characterizing the abundance and size of defects in the string-wall network. Considering also recent evidence for isotropic CMB birefringence, we find it difficult to accommodate this with the non-detection of anisotropic birefringence under the assumption that the signal is generated by an ALP defect network. △ Less

Submitted 17 August, 2022; originally announced August 2022.

Comments: 20 pages + 3 appendices, 14 figures, 3 tables

Journal ref: JCAP10(2022)090

Small-scale structure in vector dark matter

Authors: Mustafa A. Amin, Mudit Jain, Rohith Karur, Philip Mocz

Abstract: We investigate the differences in the small-scale structure of vector dark matter (VDM) and scalar dark matter (SDM) using 3+1 dimensional simulations of single/multicomponent Schrödinger-Poisson system. We find that the amount of wave interference, core-to-halo mass ratio (and its scatter), spin of the core, as well as the shape of the central regions of dark matter halos can distinguish VDM and… ▽ More We investigate the differences in the small-scale structure of vector dark matter (VDM) and scalar dark matter (SDM) using 3+1 dimensional simulations of single/multicomponent Schrödinger-Poisson system. We find that the amount of wave interference, core-to-halo mass ratio (and its scatter), spin of the core, as well as the shape of the central regions of dark matter halos can distinguish VDM and SDM. Starting with a collection of idealized halos (self-gravitating solitons) as an initial condition, we show that the system dynamically evolves to an approximately spherically symmetric configuration that has a core surrounded by a halo of interference patterns in the mass density. In the vector case, the central soliton in less dense and has a smoother transition to an $r^{-3}$ tail compared to the scalar case. As compared to SDM, wave interference in VDM is $\sim 1/\sqrt{3}$ times smaller, resulting in fewer low and high density regions, and more diffuse granules in the halo. The ratio of VDM core mass to the total halo mass is lower than that in SDM, with a steeper dependence on the total energy of the system and a slightly larger scatter. Finally, we also initiate a study of the evolution of intrinsic spin angular momentum in the VDM case. We see a positive correlation between the total intrinsic spin in the simulation and the spin of the final central core, with significant scatter. We see large intrinsic spin in the core being possible even with vanishing amounts total angular momentum in the initial conditions (at least instantaneously). Our results point towards the possibility of distinguishing VDM from SDM using astrophysical and terrestrial observations. △ Less

Submitted 17 August, 2022; v1 submitted 22 March, 2022; originally announced March 2022.

Comments: v3: 19 pages, 7 figures and an appendix. References added + minor changes to match published version

Journal ref: JCAP 08 (2022) 08, 014, 9 August 2022

Snowmass 2021 CMB-S4 White Paper

Authors: Kevork Abazajian, Arwa Abdulghafour, Graeme E. Addison, Peter Adshead, Zeeshan Ahmed, Marco Ajello, Daniel Akerib, Steven W. Allen, David Alonso, Marcelo Alvarez, Mustafa A. Amin, Mandana Amiri, Adam Anderson, Behzad Ansarinejad, Melanie Archipley, Kam S. Arnold, Matt Ashby, Han Aung, Carlo Baccigalupi, Carina Baker, Abhishek Bakshi, Debbie Bard, Denis Barkats, Darcy Barron, Peter S. Barry , et al. (331 additional authors not shown)

Abstract: This Snowmass 2021 White Paper describes the Cosmic Microwave Background Stage 4 project CMB-S4, which is designed to cross critical thresholds in our understanding of the origin and evolution of the Universe, from the highest energies at the dawn of time through the growth of structure to the present day. We provide an overview of the science case, the technical design, and project plan. This Snowmass 2021 White Paper describes the Cosmic Microwave Background Stage 4 project CMB-S4, which is designed to cross critical thresholds in our understanding of the origin and evolution of the Universe, from the highest energies at the dawn of time through the growth of structure to the present day. We provide an overview of the science case, the technical design, and project plan. △ Less

Submitted 15 March, 2022; originally announced March 2022.

Comments: Contribution to Snowmass 2021. arXiv admin note: substantial text overlap with arXiv:1908.01062, arXiv:1907.04473

Snowmass2021 Theory Frontier White Paper: Data-Driven Cosmology

Authors: Mustafa A. Amin, Francis-Yan Cyr-Racine, Tim Eifler, Raphael Flauger, Mikhail M. Ivanov, Marilena LoVerde, Caio B. de S. Nascimento, Annika H. G. Peter, Mark Vogelsberger, Scott Watson, Risa Wechsler

Abstract: Over the past few decades, astronomical and cosmological data sets firmly established the existence of physics beyond the Standard Model of particle physics by providing strong evidence for the existence of dark matter, dark energy, and non-zero neutrino mass. In addition, the generation of primordial perturbations most likely also relies on physics beyond the Standard Model of particle physics. T… ▽ More Over the past few decades, astronomical and cosmological data sets firmly established the existence of physics beyond the Standard Model of particle physics by providing strong evidence for the existence of dark matter, dark energy, and non-zero neutrino mass. In addition, the generation of primordial perturbations most likely also relies on physics beyond the Standard Model of particle physics. Theory work, ranging from models of the early universe in string theory that that led to novel phenomenological predictions to the development of effective field theories to large-scale cosmological simulations that include the physics of galaxy evolution, has played a key role in analyzing and interpreting these data sets and in suggesting novel paths forward to isolate the origins of this new physics. Over the next decade, even more sensitive surveys are beginning to take data and are being planned. In this white paper, we describe key areas of the theory program that will be needed to optimize the physics return on investment from these new observational opportunities. △ Less

Submitted 15 March, 2022; originally announced March 2022.

Comments: Contribution to Snowmass 2021

Variational Stacked Local Attention Networks for Diverse Video Captioning

Authors: Tonmoay Deb, Akib Sadmanee, Kishor Kumar Bhaumik, Amin Ahsan Ali, M Ashraful Amin, A K M Mahbubur Rahman

Abstract: While describing Spatio-temporal events in natural language, video captioning models mostly rely on the encoder's latent visual representation. Recent progress on the encoder-decoder model attends encoder features mainly in linear interaction with the decoder. However, growing model complexity for visual data encourages more explicit feature interaction for fine-grained information, which is curre… ▽ More While describing Spatio-temporal events in natural language, video captioning models mostly rely on the encoder's latent visual representation. Recent progress on the encoder-decoder model attends encoder features mainly in linear interaction with the decoder. However, growing model complexity for visual data encourages more explicit feature interaction for fine-grained information, which is currently absent in the video captioning domain. Moreover, feature aggregations methods have been used to unveil richer visual representation, either by the concatenation or using a linear layer. Though feature sets for a video semantically overlap to some extent, these approaches result in objective mismatch and feature redundancy. In addition, diversity in captions is a fundamental component of expressing one event from several meaningful perspectives, currently missing in the temporal, i.e., video captioning domain. To this end, we propose Variational Stacked Local Attention Network (VSLAN), which exploits low-rank bilinear pooling for self-attentive feature interaction and stacking multiple video feature streams in a discount fashion. Each feature stack's learned attributes contribute to our proposed diversity encoding module, followed by the decoding query stage to facilitate end-to-end diverse and natural captions without any explicit supervision on attributes. We evaluate VSLAN on MSVD and MSR-VTT datasets in terms of syntax and diversity. The CIDEr score of VSLAN outperforms current off-the-shelf methods by $7.8\%$ on MSVD and $4.5\%$ on MSR-VTT, respectively. On the same datasets, VSLAN achieves competitive results in caption diversity metrics. △ Less

Submitted 4 January, 2022; originally announced January 2022.

Comments: To be published in Winter Conference on Applications of Computer Vision 2022

Sentiment Analysis of Microblogging dataset on Coronavirus Pandemic

Authors: Nosin Ibna Mahbub, Md Rakibul Islam, Md Al Amin, Md Khairul Islam, Bikash Chandra Singh, Md Imran Hossain Showrov, Anirudda Sarkar

Abstract: Sentiment analysis can largely influence the people to get the update of the current situation. Coronavirus (COVID-19) is a contagious illness caused by the coronavirus 2 that causes severe respiratory symptoms. The lives of millions have continued to be affected by this pandemic, several countries have resorted to a full lockdown. During this lockdown, people have taken social networks to express… ▽ More Sentiment analysis can largely influence the people to get the update of the current situation. Coronavirus (COVID-19) is a contagious illness caused by the coronavirus 2 that causes severe respiratory symptoms. The lives of millions have continued to be affected by this pandemic, several countries have resorted to a full lockdown. During this lockdown, people have taken social networks to express their emotions to find a way to calm themselves down. People are spreading their sentiments through microblogging websites as one of the most preventive steps of this disease is the socialization to gain people's awareness to stay home and keep their distance when they are outside home. Twitter is a popular online social media platform for exchanging ideas. People can post their different sentiments, which can be used to aware people. But, some people want to spread fake news to frighten the people. So, it is necessary to identify the positive, negative, and neutral thoughts so that the positive opinions can be delivered to the mass people for spreading awareness to the people. Moreover, a huge volume of data is floating on Twitter. So, it is also important to identify the context of the dataset. In this paper, we have analyzed the Twitter dataset for evaluating the sentiment using several machine learning algorithms. Later, we have found out the context learning of the dataset based on the sentiments. △ Less

Submitted 17 November, 2021; originally announced November 2021.

Comments: 7 pages, 5 figures, 5th IEEE International Conference on Electrical Information and Communication Technology (EICT)

MSC Class: 68Uxx ACM Class: I.7

Journal ref: 2021 5th International Conference on Electrical Information and Communication Technology (EICT)

Polarized Vector Oscillons

Authors: Hong-Yi Zhang, Mudit Jain, Mustafa A. Amin

Abstract: Oscillons are spatially localized, time-periodic and long-lived configurations that were primarily proposed in scalar field theories with attractive self-interactions. In this letter, we demonstrate that oscillons also exist in the low-energy effective theory of an interacting massive (real) vector field. We provide two types of vector oscillons with vanishing orbital angular momentum, and approxi… ▽ More Oscillons are spatially localized, time-periodic and long-lived configurations that were primarily proposed in scalar field theories with attractive self-interactions. In this letter, we demonstrate that oscillons also exist in the low-energy effective theory of an interacting massive (real) vector field. We provide two types of vector oscillons with vanishing orbital angular momentum, and approximately spherically symmetric energy density, but not field configurations. These are: (1) "directional" oscillons (linearly polarized), with vanishing total intrinsic spin, and (2) "spinning" oscillons (circularly polarized) with a macroscopic instrinsic spin equal to $\hbar\times$ number of particles in the oscillon. In contrast to the case with only gravitational interactions, the two oscillons have different energy at a fixed particle number even in the nonrelativistic limit. By carrying out relativistic $3+1$d simulations, we show that these oscillons can be long-lived (compared to the oscillation time for the fields), and can arise from a range of Gaussian initial spatial profiles. These considerations make vector oscillons potentially relevant during the early universe and in dark photon dark matter, with novel phenomenology related to their polarization. △ Less

Submitted 23 June, 2022; v1 submitted 16 November, 2021; originally announced November 2021.

Comments: v2: 5 pages, 3 figures with appendices. Updates to appendix A and motivation, matches published version

Journal ref: Phys.Rev.D 105 (2022) 9, 096037, 27 May 2022

EuCAPT White Paper: Opportunities and Challenges for Theoretical Astroparticle Physics in the Next Decade

Authors: R. Alves Batista, M. A. Amin, G. Barenboim, N. Bartolo, D. Baumann, A. Bauswein, E. Bellini, D. Benisty, G. Bertone, P. Blasi, C. G. Böhmer, Ž. Bošnjak, T. Bringmann, C. Burrage, M. Bustamante, J. Calderón Bustillo, C. T. Byrnes, F. Calore, R. Catena, D. G. Cerdeño, S. S. Cerri, M. Chianese, K. Clough, A. Cole, P. Coloma , et al. (112 additional authors not shown)

Abstract: Astroparticle physics is undergoing a profound transformation, due to a series of extraordinary new results, such as the discovery of high-energy cosmic neutrinos with IceCube, the direct detection of gravitational waves with LIGO and Virgo, and many others. This white paper is the result of a collaborative effort that involved hundreds of theoretical astroparticle physicists and cosmologists, und… ▽ More Astroparticle physics is undergoing a profound transformation, due to a series of extraordinary new results, such as the discovery of high-energy cosmic neutrinos with IceCube, the direct detection of gravitational waves with LIGO and Virgo, and many others. This white paper is the result of a collaborative effort that involved hundreds of theoretical astroparticle physicists and cosmologists, under the coordination of the European Consortium for Astroparticle Theory (EuCAPT). Addressed to the whole astroparticle physics community, it explores upcoming theoretical opportunities and challenges for our field of research, with particular emphasis on the possible synergies among different subfields, and the prospects for solving the most fundamental open questions with multi-messenger observations. △ Less

Submitted 19 October, 2021; originally announced October 2021.

Comments: White paper of the European Consortium for Astroparticle Theory (EuCAPT). 135 authors, 400 endorsers, 133 pages, 1382 references

Polarized solitons in higher-spin wave dark matter

Authors: Mudit Jain, Mustafa A. Amin

Abstract: We first show that the effective non-relativistic theory of gravitationally interacting, massive integer-spin fields (spin-$0$, $1$, and $2$ in particular) is described by a $2s+1$ component Schrödinger-Poisson action, where $s$ is the spin of the field. We then construct $s+1$ distinct, gravitationally supported solitons in this non-relativistic theory from identically polarized plane waves. Such… ▽ More We first show that the effective non-relativistic theory of gravitationally interacting, massive integer-spin fields (spin-$0$, $1$, and $2$ in particular) is described by a $2s+1$ component Schrödinger-Poisson action, where $s$ is the spin of the field. We then construct $s+1$ distinct, gravitationally supported solitons in this non-relativistic theory from identically polarized plane waves. Such solitons are extremally polarized, with macroscopically large spin, but no orbital angular momentum. These $s+1$ solitons form a basis set, out of which partially polarized solitons can be constructed. All such solitons are ground states, have a spherically symmetric energy density but not field configurations. We discuss how solitons in higher-spin fields can be distinguished from scalar solitons, and potential gravitational and non-gravitational probes of them. △ Less

Submitted 30 November, 2021; v1 submitted 6 September, 2021; originally announced September 2021.

Comments: 14 + 2 pages, 5 figures, added more references, submitted to PRD

Journal ref: Phys. Rev. D 105, 056019, 28 March 2022

Deep Dive into Semi-Supervised ELBO for Improving Classification Performance

Authors: Fahim Faisal Niloy, M. Ashraful Amin, AKM Mahbubur Rahman, Amin Ahsan Ali

Abstract: Decomposition of the evidence lower bound (ELBO) objective of VAE used for density estimation revealed the deficiency of VAE for representation learning and suggested ways to improve the model. In this paper, we investigate whether we can get similar insights by decomposing the ELBO for semi-supervised classification using VAE model. Specifically, we show that mutual information between input and… ▽ More Decomposition of the evidence lower bound (ELBO) objective of VAE used for density estimation revealed the deficiency of VAE for representation learning and suggested ways to improve the model. In this paper, we investigate whether we can get similar insights by decomposing the ELBO for semi-supervised classification using VAE model. Specifically, we show that mutual information between input and class labels decreases during maximization of ELBO objective. We propose a method to address this issue. We also enforce cluster assumption to aid in classification. Experiments on a diverse datasets verify that our method can be used to improve the classification performance of existing VAE based semi-supervised models. Experiments also show that, this can be achieved without sacrificing the generative power of the model. △ Less

Submitted 20 November, 2022; v1 submitted 28 August, 2021; originally announced August 2021.

Comments: Under Review

A Novel Disaster Image Dataset and Characteristics Analysis using Attention Model

Authors: Fahim Faisal Niloy, Arif, Abu Bakar Siddik Nayem, Anis Sarker, Ovi Paul, M. Ashraful Amin, Amin Ahsan Ali, Moinul Islam Zaber, AKM Mahbubur Rahman

Abstract: The advancement of deep learning technology has enabled us to develop systems that outperform any other classification technique. However, success of any empirical system depends on the quality and diversity of the data available to train the proposed system. In this research, we have carefully accumulated a relatively challenging dataset that contains images collected from various sources for thr… ▽ More The advancement of deep learning technology has enabled us to develop systems that outperform any other classification technique. However, success of any empirical system depends on the quality and diversity of the data available to train the proposed system. In this research, we have carefully accumulated a relatively challenging dataset that contains images collected from various sources for three different disasters: fire, water and land. Besides this, we have also collected images for various damaged infrastructure due to natural or man made calamities and damaged human due to war or accidents. We have also accumulated image data for a class named non-damage that contains images with no such disaster or sign of damage in them. There are 13,720 manually annotated images in this dataset, each image is annotated by three individuals. We are also providing discriminating image class information annotated manually with bounding box for a set of 200 test images. Images are collected from different news portals, social media, and standard datasets made available by other researchers. A three layer attention model (TLAM) is trained and average five fold validation accuracy of 95.88% is achieved. Moreover, on the 200 unseen test images this accuracy is 96.48%. We also generate and compare attention maps for these test images to determine the characteristics of the trained attention model. Our dataset is available at https://niloy193.github.io/Disaster-Dataset △ Less

Submitted 2 July, 2021; originally announced July 2021.

Comments: ICPR 2020

Attention Toward Neighbors: A Context Aware Framework for High Resolution Image Segmentation

Authors: Fahim Faisal Niloy, M. Ashraful Amin, Amin Ahsan Ali, AKM Mahbubur Rahman

Abstract: High-resolution image segmentation remains challenging and error-prone due to the enormous size of intermediate feature maps. Conventional methods avoid this problem by using patch based approaches where each patch is segmented independently. However, independent patch segmentation induces errors, particularly at the patch boundary due to the lack of contextual information in very high-resolution… ▽ More High-resolution image segmentation remains challenging and error-prone due to the enormous size of intermediate feature maps. Conventional methods avoid this problem by using patch based approaches where each patch is segmented independently. However, independent patch segmentation induces errors, particularly at the patch boundary due to the lack of contextual information in very high-resolution images where the patch size is much smaller compared to the full image. To overcome these limitations, in this paper, we propose a novel framework to segment a particular patch by incorporating contextual information from its neighboring patches. This allows the segmentation network to see the target patch with a wider field of view without the need of larger feature maps. Comparative analysis from a number of experiments shows that our proposed framework is able to segment high resolution images with significantly improved mean Intersection over Union and overall accuracy. △ Less

Submitted 24 June, 2021; originally announced June 2021.

Comments: Accepted at ICIP 2021

Node Embedding using Mutual Information and Self-Supervision based Bi-level Aggregation

Authors: Kashob Kumar Roy, Amit Roy, A K M Mahbubur Rahman, M Ashraful Amin, Amin Ahsan Ali

Abstract: Graph Neural Networks (GNNs) learn low dimensional representations of nodes by aggregating information from their neighborhood in graphs. However, traditional GNNs suffer from two fundamental shortcomings due to their local ($l$-hop neighborhood) aggregation scheme. First, not all nodes in the neighborhood carry relevant information for the target node. Since GNNs do not exclude noisy nodes in the… ▽ More Graph Neural Networks (GNNs) learn low dimensional representations of nodes by aggregating information from their neighborhood in graphs. However, traditional GNNs suffer from two fundamental shortcomings due to their local ($l$-hop neighborhood) aggregation scheme. First, not all nodes in the neighborhood carry relevant information for the target node. Since GNNs do not exclude noisy nodes in their neighborhood, irrelevant information gets aggregated, which reduces the quality of the representation. Second, traditional GNNs also fail to capture long-range non-local dependencies between nodes. To address these limitations, we exploit mutual information (MI) to define two types of neighborhood, 1) \textit{Local Neighborhood} where nodes are densely connected within a community and each node would share higher MI with its neighbors, and 2) \textit{Non-Local Neighborhood} where MI-based node clustering is introduced to assemble informative but graphically distant nodes in the same cluster. To generate node presentations, we combine the embeddings generated by bi-level aggregation - local aggregation to aggregate features from local neighborhoods to avoid noisy information and non-local aggregation to aggregate features from non-local neighborhoods. Furthermore, we leverage self-supervision learning to estimate MI with few labeled data. Finally, we show that our model significantly outperforms the state-of-the-art methods in a wide range of assortative and disassortative graphs. △ Less

Submitted 27 April, 2021; originally announced April 2021.

Comments: Accepted at IJCNN 2021

Structure-Aware Hierarchical Graph Pooling using Information Bottleneck

Authors: Kashob Kumar Roy, Amit Roy, A K M Mahbubur Rahman, M Ashraful Amin, Amin Ahsan Ali

Abstract: Graph pooling is an essential ingredient of Graph Neural Networks (GNNs) in graph classification and regression tasks. For these tasks, different pooling strategies have been proposed to generate a graph-level representation by downsampling and summarizing nodes' features in a graph. However, most existing pooling methods are unable to capture distinguishable structural information effectively. Be… ▽ More Graph pooling is an essential ingredient of Graph Neural Networks (GNNs) in graph classification and regression tasks. For these tasks, different pooling strategies have been proposed to generate a graph-level representation by downsampling and summarizing nodes' features in a graph. However, most existing pooling methods are unable to capture distinguishable structural information effectively. Besides, they are prone to adversarial attacks. In this work, we propose a novel pooling method named as {HIBPool} where we leverage the Information Bottleneck (IB) principle that optimally balances the expressiveness and robustness of a model to learn representations of input data. Furthermore, we introduce a novel structure-aware Discriminative Pooling Readout ({DiP-Readout}) function to capture the informative local subgraph structures in the graph. Finally, our experimental results show that our model significantly outperforms other state-of-art methods on several graph classification benchmarks and more resilient to feature-perturbation attack than existing pooling methods. △ Less

Submitted 27 April, 2021; originally announced April 2021.

Comments: Accepted at IJCNN 2021

Unified Spatio-Temporal Modeling for Traffic Forecasting using Graph Neural Network

Authors: Amit Roy, Kashob Kumar Roy, Amin Ahsan Ali, M Ashraful Amin, A K M Mahbubur Rahman

Abstract: Research in deep learning models to forecast traffic intensities has gained great attention in recent years due to their capability to capture the complex spatio-temporal relationships within the traffic data. However, most state-of-the-art approaches have designed spatial-only (e.g. Graph Neural Networks) and temporal-only (e.g. Recurrent Neural Networks) modules to separately extract spatial and… ▽ More Research in deep learning models to forecast traffic intensities has gained great attention in recent years due to their capability to capture the complex spatio-temporal relationships within the traffic data. However, most state-of-the-art approaches have designed spatial-only (e.g. Graph Neural Networks) and temporal-only (e.g. Recurrent Neural Networks) modules to separately extract spatial and temporal features. However, we argue that it is less effective to extract the complex spatio-temporal relationship with such factorized modules. Besides, most existing works predict the traffic intensity of a particular time interval only based on the traffic data of the previous one hour of that day. And thereby ignores the repetitive daily/weekly pattern that may exist in the last hour of data. Therefore, we propose a Unified Spatio-Temporal Graph Convolution Network (USTGCN) for traffic forecasting that performs both spatial and temporal aggregation through direct information propagation across different timestamp nodes with the help of spectral graph convolution on a spatio-temporal graph. Furthermore, it captures historical daily patterns in previous days and current-day patterns in current-day traffic data. Finally, we validate our work's effectiveness through experimental analysis, which shows that our model USTGCN can outperform state-of-the-art performances in three popular benchmark datasets from the Performance Measurement System (PeMS). Moreover, the training time is reduced significantly with our proposed USTGCN model. △ Less

Submitted 28 April, 2021; v1 submitted 26 April, 2021; originally announced April 2021.

Comments: Accepted for publication in International Joint Conference on Neural Networks (IJCNN-2021)

Beyond Schrödinger-Poisson: Nonrelativistic Effective Field Theory for Scalar Dark Matter

Authors: Borna Salehian, Hong-Yi Zhang, Mustafa A. Amin, David I. Kaiser, Mohammad Hossein Namjoo

Abstract: Massive scalar fields provide excellent dark matter candidates, whose dynamics are often explored analytically and numerically using nonrelativistic Schrödinger-Poisson (SP) equations in a cosmological context. In this paper, starting from the nonlinear and fully relativistic Klein-Gordon-Einstein (KGE) equations in an expanding universe, we provide a systematic framework for deriving the SP equat… ▽ More Massive scalar fields provide excellent dark matter candidates, whose dynamics are often explored analytically and numerically using nonrelativistic Schrödinger-Poisson (SP) equations in a cosmological context. In this paper, starting from the nonlinear and fully relativistic Klein-Gordon-Einstein (KGE) equations in an expanding universe, we provide a systematic framework for deriving the SP equations, as well as relativistic corrections to them, by integrating out `fast modes' and including nonlinear metric and matter contributions. We provide explicit equations for the leading-order relativistic corrections, which provide insight into deviations from the SP equations as the system approaches the relativistic regime. Upon including the leading-order corrections, our equations are applicable beyond the domain of validity of the SP system, and are simpler to use than the full KGE case in some contexts. As a concrete application, we calculate the mass-radius relationship of solitons in scalar dark matter and accurately capture the deviations of this relationship from the SP system towards the KGE one. △ Less

Submitted 20 April, 2021; originally announced April 2021.

Comments: 20 pages, 5 figures, plus appendices

Report number: MIT-CTP/5293

SST-GNN: Simplified Spatio-temporal Traffic forecasting model using Graph Neural Network

Authors: Amit Roy, Kashob Kumar Roy, Amin Ahsan Ali, M Ashraful Amin, A K M Mahbubur Rahman

Abstract: To capture spatial relationships and temporal dynamics in traffic data, spatio-temporal models for traffic forecasting have drawn significant attention in recent years. Most of the recent works employed graph neural networks(GNN) with multiple layers to capture the spatial dependency. However, road junctions with different hop-distance can carry distinct traffic information which should be exploit… ▽ More To capture spatial relationships and temporal dynamics in traffic data, spatio-temporal models for traffic forecasting have drawn significant attention in recent years. Most of the recent works employed graph neural networks(GNN) with multiple layers to capture the spatial dependency. However, road junctions with different hop-distance can carry distinct traffic information which should be exploited separately but existing multi-layer GNNs are incompetent to discriminate between their impact. Again, to capture the temporal interrelationship, recurrent neural networks are common in state-of-the-art approaches that often fail to capture long-range dependencies. Furthermore, traffic data shows repeated patterns in a daily or weekly period which should be addressed explicitly. To address these limitations, we have designed a Simplified Spatio-temporal Traffic forecasting GNN(SST-GNN) that effectively encodes the spatial dependency by separately aggregating different neighborhood representations rather than with multiple layers and capture the temporal dependency with a simple yet effective weighted spatio-temporal aggregation mechanism. We capture the periodic traffic patterns by using a novel position encoding scheme with historical and current data in two different models. With extensive experimental analysis, we have shown that our model has significantly outperformed the state-of-the-art models on three real-world traffic datasets from the Performance Measurement System (PeMS). △ Less

Submitted 31 March, 2021; originally announced April 2021.

Comments: Accepted for publication in 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD-2021)

Dipole Radiation and Beyond from Axion Stars in Electromagnetic Fields

Authors: Mustafa A. Amin, Andrew J. Long, Zong-Gang Mou, Paul Saffin

Abstract: We investigate the production of photons from coherently oscillating, spatially localized clumps of axionic fields (oscillons and axion stars) in the presence of external electromagnetic fields. We delineate different qualitative behaviour of the photon luminosity in terms of an effective dimensionless coupling parameter constructed out of the axion-photon coupling, and field amplitude, oscillatio… ▽ More We investigate the production of photons from coherently oscillating, spatially localized clumps of axionic fields (oscillons and axion stars) in the presence of external electromagnetic fields. We delineate different qualitative behaviour of the photon luminosity in terms of an effective dimensionless coupling parameter constructed out of the axion-photon coupling, and field amplitude, oscillation frequency and radius of the axion star. For small values of this dimensionless coupling, we provide a general analytic formula for the dipole radiation field and the photon luminosity per solid angle, including a strong dependence on the radius of the configuration. For moderate to large coupling, we report on a non-monotonic behavior of the luminosity with the coupling strength in the presence of external magnetic fields. After an initial rise in luminosity with the coupling strength, we see a suppression (by an order of magnitude or more compared to the dipole radiation approximation) at moderately large coupling. At sufficiently large coupling, we find a transition to a regime of exponential growth of the luminosity due to parametric resonance. We carry out 3+1 dimensional lattice simulations of axion electrodynamics, at small and large coupling, including non-perturbative effects of parametric resonance as well as backreaction effects when necessary. We also discuss medium (plasma) effects that lead to resonant axion to photon conversion, relevance of the coherence of the soliton, and implications of our results in astrophysical and cosmological settings. △ Less

Submitted 22 March, 2021; originally announced March 2021.

Comments: 12 figures, 28 pages + 1 Appendix

CMB birefringence from ultra-light axion string networks

Authors: Mudit Jain, Andrew J. Long, Mustafa A. Amin

Abstract: The polarization of Cosmic Microwave Background (CMB) photons is rotated as they pass through (ultralight-) axion string loops. Studying this birefringence can reveal valuable information about the axion-photon coupling and the structure of the string network. We develop an approximate analytic formalism and identify a kernel function that can be used to calculate the two-point correlation functio… ▽ More The polarization of Cosmic Microwave Background (CMB) photons is rotated as they pass through (ultralight-) axion string loops. Studying this birefringence can reveal valuable information about the axion-photon coupling and the structure of the string network. We develop an approximate analytic formalism and identify a kernel function that can be used to calculate the two-point correlation function for CMB birefringence induced by an arbitrary axion string network. Using this formalism, we evaluate the birefringence signal for some simple loop distributions (including scaling and network collapse). We find that the angular correlation function has a characteristic angular scale set by $θ_\mathrm{min}$, which corresponds to the angular extent of the loops at the time of recombination. This results in a peak in the birefringence power spectrum around $\ell_p \sim 1/θ_\mathrm{min}$. An additional scale, controlled by the axion's mass, is introduced if the network collapses before today. △ Less

Submitted 21 May, 2021; v1 submitted 19 March, 2021; originally announced March 2021.

Comments: 31 pages + appendices; 9 figures. v2: updated towards the version published in JCAP (added limits from ACTPol n SPTPol in power spectrum figures; added more references)

Journal ref: JCAP05(2021)055

Hierarchical Self Attention Based Autoencoder for Open-Set Human Activity Recognition

Authors: M Tanjid Hasan Tonmoy, Saif Mahmud, A K M Mahbubur Rahman, M Ashraful Amin, Amin Ahsan Ali

Abstract: Wearable sensor based human activity recognition is a challenging problem due to difficulty in modeling spatial and temporal dependencies of sensor signals. Recognition models in closed-set assumption are forced to yield members of known activity classes as prediction. However, activity recognition models can encounter an unseen activity due to body-worn sensor malfunction or disability of the sub… ▽ More Wearable sensor based human activity recognition is a challenging problem due to difficulty in modeling spatial and temporal dependencies of sensor signals. Recognition models in closed-set assumption are forced to yield members of known activity classes as prediction. However, activity recognition models can encounter an unseen activity due to body-worn sensor malfunction or disability of the subject performing the activities. This problem can be addressed through modeling solution according to the assumption of open-set recognition. Hence, the proposed self attention based approach combines data hierarchically from different sensor placements across time to classify closed-set activities and it obtains notable performance improvement over state-of-the-art models on five publicly available datasets. The decoder in this autoencoder architecture incorporates self-attention based feature representations from encoder to detect unseen activity classes in open-set recognition setting. Furthermore, attention maps generated by the hierarchical model demonstrate explainable selection of features in activity recognition. We conduct extensive leave one subject out validation experiments that indicate significantly improved robustness to noise and subject specific variability in body-worn sensor signals. The source code is available at: github.com/saif-mahmud/hierarchical-attention-HAR △ Less

Submitted 7 March, 2021; originally announced March 2021.

Comments: Accepted for publication in 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD-2021)

Deep-learning coupled with novel classification method to classify the urban environment of the developing world

Authors: Qianwei Cheng, AKM Mahbubur Rahman, Anis Sarker, Abu Bakar Siddik Nayem, Ovi Paul, Amin Ahsan Ali, M Ashraful Amin, Ryosuke Shibasaki, Moinul Zaber

Abstract: Rapid globalization and the interdependence of humanity that engender tremendous in-flow of human migration towards the urban spaces. With advent of high definition satellite images, high resolution data, computational methods such as deep neural network, capable hardware; urban planning is seeing a paradigm shift. Legacy data on urban environments are now being complemented with high-volume, high… ▽ More Rapid globalization and the interdependence of humanity that engender tremendous in-flow of human migration towards the urban spaces. With advent of high definition satellite images, high resolution data, computational methods such as deep neural network, capable hardware; urban planning is seeing a paradigm shift. Legacy data on urban environments are now being complemented with high-volume, high-frequency data. In this paper we propose a novel classification method that is readily usable for machine analysis and show applicability of the methodology on a developing world setting. The state-of-the-art is mostly dominated by classification of building structures, building types etc. and largely represents the developed world which are insufficient for developing countries such as Bangladesh where the surrounding is crucial for the classification. Moreover, the traditional methods propose small-scale classifications, which give limited information with poor scalability and are slow to compute. We categorize the urban area in terms of informal and formal spaces taking the surroundings into account. 50 km x 50 km Google Earth image of Dhaka, Bangladesh was visually annotated and categorized by an expert. The classification is based broadly on two dimensions: urbanization and the architectural form of urban environment. Consequently, the urban space is divided into four classes: 1) highly informal; 2) moderately informal; 3) moderately formal; and 4) highly formal areas. In total 16 sub-classes were identified. For semantic segmentation, Google's DeeplabV3+ model was used which increases the field of view of the filters to incorporate larger context. Image encompassing 70% of the urban space was used for training and the remaining 30% was used for testing and validation. The model is able to segment with 75% accuracy and 60% Mean IoU. △ Less

Submitted 7 January, 2021; v1 submitted 25 November, 2020; originally announced November 2020.

Comments: Accepted paper at 2nd International Conference on Signal Processing and Machine Learning (SIGML 2021); 20 pages, 7 figures, 1 table

Modeling Traffic Congestion in Developing Countries using Google Maps Data

Authors: Md. Aktaruzzaman Pramanik, Md Mahbubur Rahman, ASM Iftekhar Anam, Amin Ahsan Ali, M Ashraful Amin, A K M Mahbubur Rahman

Abstract: Traffic congestion research is on the rise, thanks to urbanization, economic growth, and industrialization. Developed countries invest a lot of research money in collecting traffic data using Radio Frequency Identification (RFID), loop detectors, speed sensors, high-end traffic light, and GPS. However, these processes are expensive, infeasible, and non-scalable for developing countries with numero… ▽ More Traffic congestion research is on the rise, thanks to urbanization, economic growth, and industrialization. Developed countries invest a lot of research money in collecting traffic data using Radio Frequency Identification (RFID), loop detectors, speed sensors, high-end traffic light, and GPS. However, these processes are expensive, infeasible, and non-scalable for developing countries with numerous non-motorized vehicles, proliferated ride-sharing services, and frequent pedestrians. This paper proposes a novel approach to collect traffic data from Google Map's traffic layer with minimal cost. We have implemented widely used models such as Historical Averages (HA), Support Vector Regression (SVR), Support Vector Regression with Graph (SVR-Graph), Auto-Regressive Integrated Moving Average (ARIMA) to show the efficacy of the collected traffic data in forecasting future congestion. We show that even with these simple models, we could predict the traffic congestion ahead of time. We also demonstrate that the traffic patterns are significantly different between weekdays and weekends. △ Less

Submitted 29 October, 2020; originally announced November 2020.

Electromagnetic Bursts from Mergers of Oscillons in Axion-like Fields

Authors: Mustafa A. Amin, Zong-Gang Mou

Abstract: We investigate the bursts of electromagnetic and scalar radiation resulting from the collision, and merger of oscillons made from axion-like particles using 3+1 dimensional lattice simulations of the coupled axion-gauge field system. The radiation into photons is suppressed before the merger. However, it becomes the dominant source of energy loss after the merger if a resonance condition is satisf… ▽ More We investigate the bursts of electromagnetic and scalar radiation resulting from the collision, and merger of oscillons made from axion-like particles using 3+1 dimensional lattice simulations of the coupled axion-gauge field system. The radiation into photons is suppressed before the merger. However, it becomes the dominant source of energy loss after the merger if a resonance condition is satisfied. Conversely, the radiation in scalar waves is dominant during initial merger phase but suppressed after the merger. The backreaction of scalar and electromagnetic radiation is included in our simulations. We evolve the system long enough to see that the resonant photon production extracts a significant fraction of the initial axion energy, and again falls out of the resonance condition. We provide a parametric understanding of the time, and energy scales involved in the process and discuss observational prospects of detecting the electromagnetic signal. △ Less

Submitted 13 October, 2020; v1 submitted 23 September, 2020; originally announced September 2020.

Comments: 21 pages, 6 figures + 4 pages of Appendix (v2, updated references and minor changes in text)

CMB-S4: Forecasting Constraints on Primordial Gravitational Waves

Authors: CMB-S4 Collaboration, :, Kevork Abazajian, Graeme E. Addison, Peter Adshead, Zeeshan Ahmed, Daniel Akerib, Aamir Ali, Steven W. Allen, David Alonso, Marcelo Alvarez, Mustafa A. Amin, Adam Anderson, Kam S. Arnold, Peter Ashton, Carlo Baccigalupi, Debbie Bard, Denis Barkats, Darcy Barron, Peter S. Barry, James G. Bartlett, Ritoban Basu Thakur, Nicholas Battaglia, Rachel Bean, Chris Bebek , et al. (212 additional authors not shown)

Abstract: CMB-S4---the next-generation ground-based cosmic microwave background (CMB) experiment---is set to significantly advance the sensitivity of CMB measurements and enhance our understanding of the origin and evolution of the Universe, from the highest energies at the dawn of time through the growth of structure to the present day. Among the science cases pursued with CMB-S4, the quest for detecting p… ▽ More CMB-S4---the next-generation ground-based cosmic microwave background (CMB) experiment---is set to significantly advance the sensitivity of CMB measurements and enhance our understanding of the origin and evolution of the Universe, from the highest energies at the dawn of time through the growth of structure to the present day. Among the science cases pursued with CMB-S4, the quest for detecting primordial gravitational waves is a central driver of the experimental design. This work details the development of a forecasting framework that includes a power-spectrum-based semi-analytic projection tool, targeted explicitly towards optimizing constraints on the tensor-to-scalar ratio, $r$, in the presence of Galactic foregrounds and gravitational lensing of the CMB. This framework is unique in its direct use of information from the achieved performance of current Stage 2--3 CMB experiments to robustly forecast the science reach of upcoming CMB-polarization endeavors. The methodology allows for rapid iteration over experimental configurations and offers a flexible way to optimize the design of future experiments given a desired scientific goal. To form a closed-loop process, we couple this semi-analytic tool with map-based validation studies, which allow for the injection of additional complexity and verification of our forecasts with several independent analysis methods. We document multiple rounds of forecasts for CMB-S4 using this process and the resulting establishment of the current reference design of the primordial gravitational-wave component of the Stage-4 experiment, optimized to achieve our science goals of detecting primordial gravitational waves for $r > 0.003$ at greater than $5σ$, or, in the absence of a detection, of reaching an upper limit of $r < 0.001$ at $95\%$ CL. △ Less

Submitted 27 August, 2020; originally announced August 2020.

Comments: 24 pages, 8 figures, 9 tables, submitted to ApJ. arXiv admin note: text overlap with arXiv:1907.04473

LULC Segmentation of RGB Satellite Image Using FCN-8

Authors: Abu Bakar Siddik Nayem, Anis Sarker, Ovi Paul, Amin Ali, Md. Ashraful Amin, AKM Mahbubur Rahman

Abstract: This work presents use of Fully Convolutional Network (FCN-8) for semantic segmentation of high-resolution RGB earth surface satel-lite images into land use land cover (LULC) categories. Specically, we propose a non-overlapping grid-based approach to train a Fully Convo-lutional Network (FCN-8) with vgg-16 weights to segment satellite im-ages into four (forest, built-up, farmland and water) classe… ▽ More This work presents use of Fully Convolutional Network (FCN-8) for semantic segmentation of high-resolution RGB earth surface satel-lite images into land use land cover (LULC) categories. Specically, we propose a non-overlapping grid-based approach to train a Fully Convo-lutional Network (FCN-8) with vgg-16 weights to segment satellite im-ages into four (forest, built-up, farmland and water) classes. The FCN-8 semantically projects the discriminating features in lower resolution learned by the encoder onto the pixel space in higher resolution to get a dense classi cation. We experimented the proposed system with Gaofen-2 image dataset, that contains 150 images of over 60 di erent cities in china. For comparison, we used available ground-truth along with images segmented using a widely used commeriial GIS software called eCogni-tion. With the proposed non-overlapping grid-based approach, FCN-8 obtains signi cantly improved performance, than the eCognition soft-ware. Our model achieves average accuracy of 91.0% and average Inter-section over Union (IoU) of 0.84. In contrast, eCognitions average accu-racy is 74.0% and IoU is 0.60. This paper also reports a detail analysis of errors occurred at the LULC boundary. △ Less

Submitted 24 August, 2020; originally announced August 2020.

Comments: Accepted paper at 3rd SLAAI-International Conference on Artificial Intelligence; 13 pages, 7 figures, 3 tables

The First Three Seconds: a Review of Possible Expansion Histories of the Early Universe

Authors: Rouzbeh Allahverdi, Mustafa A. Amin, Asher Berlin, Nicolás Bernal, Christian T. Byrnes, M. Sten Delos, Adrienne L. Erickcek, Miguel Escudero, Daniel G. Figueroa, Katherine Freese, Tomohiro Harada, Dan Hooper, David I. Kaiser, Tanvi Karwal, Kazunori Kohri, Gordan Krnjaic, Marek Lewicki, Kaloian D. Lozanov, Vivian Poulin, Kuver Sinha, Tristan L. Smith, Tomo Takahashi, Tommi Tenkanen, James Unwin, Ville Vaskonen , et al. (1 additional authors not shown)

Abstract: It is commonly assumed that the energy density of the Universe was dominated by radiation between reheating after inflation and the onset of matter domination 54,000 years later. While the abundance of light elements indicates that the Universe was radiation dominated during Big Bang Nucleosynthesis (BBN), there is scant evidence that the Universe was radiation dominated prior to BBN. It is theref… ▽ More It is commonly assumed that the energy density of the Universe was dominated by radiation between reheating after inflation and the onset of matter domination 54,000 years later. While the abundance of light elements indicates that the Universe was radiation dominated during Big Bang Nucleosynthesis (BBN), there is scant evidence that the Universe was radiation dominated prior to BBN. It is therefore possible that the cosmological history was more complicated, with deviations from the standard radiation domination during the earliest epochs. Indeed, several interesting proposals regarding various topics such as the generation of dark matter, matter-antimatter asymmetry, gravitational waves, primordial black holes, or microhalos during a nonstandard expansion phase have been recently made. In this paper, we review various possible causes and consequences of deviations from radiation domination in the early Universe - taking place either before or after BBN - and the constraints on them, as they have been discussed in the literature during the recent years. △ Less

Submitted 27 January, 2021; v1 submitted 29 June, 2020; originally announced June 2020.

Comments: 67 pages, 18 figures. v2: Discussion and references added. Accepted for publication in The Open Journal of Astrophysics

Report number: FERMILAB-PUB-20-242-A, KCL-PH-TH/2020-33, KEK-Cosmo-257, KEK-TH-2231, IPMU20-0070, PI/UAN-2020-674FT, RUP-20-22

Journal ref: Open J. Astrophys. Vol. 4, 2021

Classical Decay Rates of Oscillons

Authors: Hong-Yi Zhang, Mustafa A. Amin, Edmund J. Copeland, Paul M. Saffin, Kaloian D. Lozanov

Abstract: Oscillons are extremely long-lived, spatially-localized field configurations in real-valued scalar field theories that slowly lose energy via radiation of scalar waves. Before their eventual demise, oscillons can pass through (one or more) exceptionally stable field configurations where their decay rate is highly suppressed. We provide an improved calculation of the non-trivial behavior of the dec… ▽ More Oscillons are extremely long-lived, spatially-localized field configurations in real-valued scalar field theories that slowly lose energy via radiation of scalar waves. Before their eventual demise, oscillons can pass through (one or more) exceptionally stable field configurations where their decay rate is highly suppressed. We provide an improved calculation of the non-trivial behavior of the decay rates, and lifetimes of oscillons. In particular, our calculation correctly captures the existence (or absence) of the exceptionally long-lived states for large amplitude oscillons in a broad class of potentials, including non-polynomial potentials that flatten at large field values. The key underlying reason for the improved (by many orders of magnitude in some cases) calculation is the systematic inclusion of a spacetime-dependent effective mass term in the equation describing the radiation emitted by oscillons (in addition to a source term). Our results for the exceptionally stable configurations, decay rates, and lifetime of large amplitude oscillons (in some cases $\gtrsim 10^8$ oscillations) in such flattened potentials might be relevant for cosmological applications. △ Less

Submitted 19 May, 2020; v1 submitted 2 April, 2020; originally announced April 2020.

Comments: v2: 27 pages, 7 figures. References added, added short discussion on cosine potentials

Human Activity Recognition from Wearable Sensor Data Using Self-Attention

Authors: Saif Mahmud, M Tanjid Hasan Tonmoy, Kishor Kumar Bhaumik, A K M Mahbubur Rahman, M Ashraful Amin, Mohammad Shoyaib, Muhammad Asif Hossain Khan, Amin Ahsan Ali

Abstract: Human Activity Recognition from body-worn sensor data poses an inherent challenge in capturing spatial and temporal dependencies of time-series signals. In this regard, the existing recurrent or convolutional or their hybrid models for activity recognition struggle to capture spatio-temporal context from the feature space of sensor reading sequence. To address this complex problem, we propose a se… ▽ More Human Activity Recognition from body-worn sensor data poses an inherent challenge in capturing spatial and temporal dependencies of time-series signals. In this regard, the existing recurrent or convolutional or their hybrid models for activity recognition struggle to capture spatio-temporal context from the feature space of sensor reading sequence. To address this complex problem, we propose a self-attention based neural network model that foregoes recurrent architectures and utilizes different types of attention mechanisms to generate higher dimensional feature representation used for classification. We performed extensive experiments on four popular publicly available HAR datasets: PAMAP2, Opportunity, Skoda and USC-HAD. Our model achieve significant performance improvement over recent state-of-the-art models in both benchmark test subjects and Leave-one-subject-out evaluation. We also observe that the sensor attention maps produced by our model is able capture the importance of the modality and placement of the sensors in predicting the different activity classes. △ Less

Submitted 17 March, 2020; originally announced March 2020.

Comments: Accepted for publication at the 24th European Conference on Artificial Intelligence (ECAI-2020); 8 pages, 4 figures

Curvature Perturbations From Stochastic Particle Production During Inflation

Authors: Marcos A. G. Garcia, Mustafa A. Amin, Daniel Green

Abstract: We calculate the curvature power spectrum sourced by spectator fields that are excited repeatedly and non-adiabatically during inflation. In the absence of detailed information of the nature of spectator field interactions, we consider an ensemble of models with intervals between the repeated interactions and interaction strengths drawn from simple probabilistic distributions. We show that the cur… ▽ More We calculate the curvature power spectrum sourced by spectator fields that are excited repeatedly and non-adiabatically during inflation. In the absence of detailed information of the nature of spectator field interactions, we consider an ensemble of models with intervals between the repeated interactions and interaction strengths drawn from simple probabilistic distributions. We show that the curvature power spectra of each member of the ensemble shows rich structure with many features, and there is a large variability between different realizations of the same ensemble. Such features can be probed by the cosmic microwave background (CMB) and large scale structure observations. They can also have implications for primordial black hole formation and CMB spectral distortions. The geometric random walk behavior of the spectator field allows us to calculate the ensemble-averaged power spectrum of curvature perturbations semi-analytically. For sufficiently large stochastic sourcing, the ensemble-averaged power spectrum shows a scale dependence arising from the time spent by modes outside the horizon during the period of particle production, in spite of there being no preferred scale in the underlying model. We find that the magnitude of the ensemble-averaged power spectrum overestimates the typical power spectra in the ensemble because the ensemble distribution of the power spectra is highly non-Gaussian with fat tails. △ Less

Submitted 24 January, 2020; originally announced January 2020.

Comments: 68 pages, 23 figures

Report number: IFT-UAM/CSIC-20-14

GFiRe: a Gauge Field integrator for Reheating

Authors: Kaloian D. Lozanov, Mustafa A. Amin

Abstract: We present a new numerical algorithm and code, ${\sf GFiRe}$, for solving the non-linear evolution of Abelian gauge fields coupled to complex scalar fields in homogeneous and isotropic spacetimes. We adopt a hybrid approach to solving the system: the spatial derivatives are discretized using standard Lattice Gauge Field Theory techniques, whereas the time evolution of the fields and scale factor i… ▽ More We present a new numerical algorithm and code, ${\sf GFiRe}$, for solving the non-linear evolution of Abelian gauge fields coupled to complex scalar fields in homogeneous and isotropic spacetimes. We adopt a hybrid approach to solving the system: the spatial derivatives are discretized using standard Lattice Gauge Field Theory techniques, whereas the time evolution of the fields and scale factor is implemented with explicit, composite, symplectic integrators. An important property of our compound algorithm is that the discretized Gauss constraint is respected exactly, regardless of the order of the symplectic integrator. This remains true even when the background expansion is computed "self-consistently"; that is, when the expansion history is computed using spatial averaged components of the energy momentum tensor in the simulation volume. Hence, our code can also be used in cases where the fields dominate the energy density of the universe, for example, during reheating after inflation. We test the algorithm in scenarios of reheating where the inflaton is a complex scalar field with a potential $\propto(2|\varphi|^2-v^2)^2$ and is coupled to an Abelian gauge field. Tracing the evolution of the system through complex dynamics (including resonant excitation of fields, backreaction, formation of solitons, and changes in the equation of state) in a self-consistently expanding universe, we find the energy conservation violation ($<10^{-4}$) to be very stable and the Gauss constraint violation ($<10^{-6}$) to be dominated by differencing noise. △ Less

Submitted 26 February, 2020; v1 submitted 15 November, 2019; originally announced November 2019.

Comments: 35 pages, 13 figures; minor changes, accepted for publication in JCAP

Galaxy Formation with BECDM -- II. Cosmic Filaments and First Galaxies

Authors: Philip Mocz, Anastasia Fialkov, Mark Vogelsberger, Fernando Becerra, Xuejian Shen, Victor H. Robles, Mustafa A. Amin, Jesús Zavala, Michael Boylan-Kolchin, Sownak Bose, Federico Marinacci, Pierre-Henri Chavanis, Lachlan Lancaster, Lars Hernquist

Abstract: Bose-Einstein Condensate Dark Matter (BECDM; also known as Fuzzy Dark Matter) is motivated by fundamental physics and has recently received significant attention as a serious alternative to the established Cold Dark Matter (CDM) model. We perform cosmological simulations of BECDM gravitationally coupled to baryons and investigate structure formation at high redshifts ($z \gtrsim 5$) for a boson ma… ▽ More Bose-Einstein Condensate Dark Matter (BECDM; also known as Fuzzy Dark Matter) is motivated by fundamental physics and has recently received significant attention as a serious alternative to the established Cold Dark Matter (CDM) model. We perform cosmological simulations of BECDM gravitationally coupled to baryons and investigate structure formation at high redshifts ($z \gtrsim 5$) for a boson mass $m=2.5\cdot 10^{-22}~{\rm eV}$, exploring the dynamical effects of its wavelike nature on the cosmic web and the formation of first galaxies. Our BECDM simulations are directly compared to CDM as well as to simulations where the dynamical quantum potential is ignored and only the initial suppression of the power spectrum is considered -- a Warm Dark Matter-like ("WDM") model often used as a proxy for BECDM. Our simulations confirm that "WDM" is a good approximation to BECDM on large cosmological scales even in the presence of the baryonic feedback. Similarities also exist on small scales, with primordial star formation happening both in isolated haloes and continuously along cosmic filaments; the latter effect is not present in CDM. Global star formation and metal enrichment in these first galaxies are delayed in BECDM/"WDM" compared to the CDM case: in BECDM/"WDM" first stars form at $z\sim 13$/$13.5$ while in CDM star formation starts at $z\sim 35$. The signature of BECDM interference, not present in "WDM", is seen in the evolved dark matter power spectrum: although the small scale structure is initially suppressed, power on kpc scales is added at lower redshifts. Our simulations lay the groundwork for realistic simulations of galaxy formation in BECDM. △ Less

Submitted 13 November, 2019; originally announced November 2019.

Comments: 19 pages, 12 figures, submitted to MNRAS

First star-forming structures in fuzzy cosmic filaments

Authors: Philip Mocz, Anastasia Fialkov, Mark Vogelsberger, Fernando Becerra, Mustafa A. Amin, Sownak Bose, Michael Boylan-Kolchin, Pierre-Henri Chavanis, Lars Hernquist, Lachlan Lancaster, Federico Marinacci, Victor Robles, Jesús Zavala

Abstract: In hierarchical models of structure formation, the first galaxies form in low-mass dark matter potential wells, probing the behavior of dark matter on kiloparsec (kpc) scales. Even though these objects are below the detection threshold of current telescopes, future missions will open an observational window into this emergent world. In this Letter we investigate how the first galaxies are assemble… ▽ More In hierarchical models of structure formation, the first galaxies form in low-mass dark matter potential wells, probing the behavior of dark matter on kiloparsec (kpc) scales. Even though these objects are below the detection threshold of current telescopes, future missions will open an observational window into this emergent world. In this Letter we investigate how the first galaxies are assembled in a `fuzzy' dark matter (FDM) cosmology where dark matter is an ultralight $\sim 10^{-22}$~eV boson and the primordial stars are expected to form along dense dark matter filaments. Using a first-of-its-kind cosmological hydrodynamical simulation, we explore the interplay between baryonic physics and unique wavelike features inherent to FDM. In our simulation, the dark matter filaments show coherent interference patterns on the boson de Broglie scale and develop cylindrical soliton-like cores which are unstable under gravity and collapse into kpc-scale spherical solitons. Features of the dark matter distribution are largely unaffected by the baryonic feedback. On the contrary, the distributions of gas and stars, which do form along the entire filament, exhibit central cores imprinted by dark matter -- a smoking gun signature of FDM. △ Less

Submitted 7 October, 2019; v1 submitted 3 October, 2019; originally announced October 2019.

Comments: 6 pages, 2 figures, accepted to Phys. Rev. Lett

Oscillating scalar fields and the Hubble tension: a resolution with novel signatures

Authors: Tristan L. Smith, Vivian Poulin, Mustafa A. Amin

Abstract: We present a detailed investigation of a sub-dominant oscillating scalar field ('early dark energy', EDE) in the context of resolving the Hubble tension. Consistent with earlier work, but without relying on fluid approximations, we find that a scalar field frozen due to Hubble friction until ${\rm log}_{10}(z_c)\sim3.5$, reaching $ρ_{\rm EDE}(z_c)/ρ_{\rm tot}\sim10$%, and diluting faster than matt… ▽ More We present a detailed investigation of a sub-dominant oscillating scalar field ('early dark energy', EDE) in the context of resolving the Hubble tension. Consistent with earlier work, but without relying on fluid approximations, we find that a scalar field frozen due to Hubble friction until ${\rm log}_{10}(z_c)\sim3.5$, reaching $ρ_{\rm EDE}(z_c)/ρ_{\rm tot}\sim10$%, and diluting faster than matter afterwards can bring cosmic microwave background (CMB), baryonic acoustic oscillations, supernovae luminosity distances, and the late-time estimate of the Hubble constant from the SH0ES collaboration into agreement. A scalar field potential which scales as $V(φ) \propto φ^{2n}$ with $2\lesssim n\lesssim 3.4$ around the minimum is preferred at the 68% confidence level, and the {\em Planck} polarization places additional constraints on the dynamics of perturbations in the scalar field. In particular, the data prefers a potential which flattens at large field displacements. An MCMC analysis of mock data shows that the next-generation CMB observations (i.e., CMB-S4) can unambiguously detect the presence of the EDE at very high significance. This projected sensitivity to the EDE dynamics is mainly driven by improved measurements of the $E$-mode polarization. We also explore new observational signatures of EDE scalar field dynamics: (i) We find that depending on the strength of the tensor-to-scalar ratio, the presence of the EDE might imply the existence of isocurvature perturbations in the CMB. (ii) We show that a strikingly rapid, scale-dependent growth of EDE field perturbations can result from parametric resonance driven by the anharmonic oscillating field for $n\approx 2$. This instability and ensuing potentially nonlinear, spatially inhomogenoues, dynamics may provide unique signatures of this scenario. △ Less

Submitted 19 August, 2019; originally announced August 2019.

Comments: 26 pages, 22 figures, comments welcome

Journal ref: Phys. Rev. D 101, 063523 (2020)