About
Activity
-
I enjoyed hanging out with Pika founder Demi G. (Neo Scholar '17). 🤩 College CS students, apply to be a Neo Scholar today! P.S. Apply by tomorrow…
I enjoyed hanging out with Pika founder Demi G. (Neo Scholar '17). 🤩 College CS students, apply to be a Neo Scholar today! P.S. Apply by tomorrow…
Liked by Nancy Xu
-
This map gives me chills. A year ago this was a school project working with Nick Bloom -- literally "Project Pair" -- using whiteboards and sticky…
This map gives me chills. A year ago this was a school project working with Nick Bloom -- literally "Project Pair" -- using whiteboards and sticky…
Liked by Nancy Xu
-
Close to half of the kids in Ukraine are unable to go to full-time school. One of those kids is Nazar Drugov, who studied using Khan Academy during…
Close to half of the kids in Ukraine are unable to go to full-time school. One of those kids is Nazar Drugov, who studied using Khan Academy during…
Liked by Nancy Xu
Experience & Education
Licenses & Certifications
Volunteer Experience
-
Founder, Editor in Chief
The Gradient
- Present 7 years
Join 100k+ AI industry leaders and researchers who subscribe to the latest in all things AI startups, research, and more. https://thegradient.pub/
-
President (Board of Directors)
Stanford Women in Computer Science
- 2 years
Education
• Lead 700+ person community and board.
Outreach and high-profile CS industry events for the Stanford community
• Manage initiatives between Stanford WiCS and industry partners. -
Stanford President
ACM, Association for Computing Machinery
- 1 year
Science and Technology
(Stanford ACM)
-
Publications
-
The Kipoi repository accelerates community exchange and reuse of predictive models for genomics
Nature Biotechnology
Advances in machine learning, coupled with rapidly growing genome sequencing and molecular profiling datasets, are catalyzing progress in genomics1. In particular, predictive machine learning models, which are mathematical functions trained to map input data to output values, have found widespread usage. Prominent examples include calling variants from whole-genome sequencing data2,3, estimating CRISPR guide activity4,5 and predicting molecular phenotypes, including transcription factor…
Advances in machine learning, coupled with rapidly growing genome sequencing and molecular profiling datasets, are catalyzing progress in genomics1. In particular, predictive machine learning models, which are mathematical functions trained to map input data to output values, have found widespread usage. Prominent examples include calling variants from whole-genome sequencing data2,3, estimating CRISPR guide activity4,5 and predicting molecular phenotypes, including transcription factor binding, chromatin accessibility and splicing efficiency, from DNA sequence1,6,7,8,9,10,11. Once trained, these models can be probed in silico to infer quantitative relationships between diverse genomic data modalities, enabling several key applications such as the interpretation of functional genetic variants and rational design of synthetic genes.
-
Graph isomorphisms: On Furst-Hopcroft-Luks algorithm for group order
Stanford University Mathematics Department
Groups and their derivative properties arise naturally in models of the world — bridging
intuition for graph symmetries and other mathematical structures. This presentation will
explore the question of how we can derive the order of a group given a set of its generators.
We show an algorithmic result by Furst-Hopcroft-Luks for finding the order of a symmetric
group subgroup in polynomial time. -
Deep Reinforcement Learning for Simulated Autonomous Driving
Stanford University
This research explores deep Q-learning for autonomous driving in The Open Racing Car Simulator (TORCS). Using the TensorFlow and Keras software frameworks, we train fully-connected deep neural networks that are able to autonomously drive across a diverse range of track geometries. An initial proof-of-concept of classical Q-learning was implemented in Flappy Bird. A reward function promoting longitudinal velocity while penalizing transverse velocity and divergence from the track center is used…
This research explores deep Q-learning for autonomous driving in The Open Racing Car Simulator (TORCS). Using the TensorFlow and Keras software frameworks, we train fully-connected deep neural networks that are able to autonomously drive across a diverse range of track geometries. An initial proof-of-concept of classical Q-learning was implemented in Flappy Bird. A reward function promoting longitudinal velocity while penalizing transverse velocity and divergence from the track center is used to train the agent. To validate learning, the research analyzes the reward function parameters of the models over two validation tracks and qualitatively assesses driving stability. A video of the learned agent driving in TORCS can be found online: https://www. dropbox. com/sh/b4623soznsjmp12/AAA1UD8_oaa94FgFC6eyxReya? dl= 0.
-
Grounding Open-Domain Instructions to Automate Web Support Tasks
North American Association for Computational Linguistics Conference (NAACL)
Grounding natural language instructions on the web to perform previously unseen tasks enables accessibility and automation. We introduce a task and dataset to train AI agents from open-domain, step-by-step instructions originally written for people. We build RUSS (Rapid Universal Support Service) to tackle this problem. RUSS consists of two models: First, a BERT-LSTM with pointers parses instructions to ThingTalk, a domain-specific language we design for grounding natural language on the web���
Grounding natural language instructions on the web to perform previously unseen tasks enables accessibility and automation. We introduce a task and dataset to train AI agents from open-domain, step-by-step instructions originally written for people. We build RUSS (Rapid Universal Support Service) to tackle this problem. RUSS consists of two models: First, a BERT-LSTM with pointers parses instructions to ThingTalk, a domain-specific language we design for grounding natural language on the web. Then, a grounding model retrieves the unique IDs of any webpage elements requested in ThingTalk. RUSS may interact with the user through a dialogue (e.g. ask for an address) or execute a web operation (e.g. click a button) inside the web runtime. To augment training, we synthesize natural language instructions mapped to ThingTalk. Our dataset consists of 80 different customer service problems from help websites, with a total of 741 step-by-step instructions and their corresponding actions. RUSS achieves 76.7% end-to-end accuracy predicting agent actions from single instructions. It outperforms state-of-the-art models that directly map instructions to actions without ThingTalk. Our user study shows that RUSS is preferred by actual users over web navigation.
-
An Analysis of the Mathematical Structure of Simpson's Paradox with Applications
Parabola
This paper analyzes the mathematical foundations of Simpson's Paradox and proposes two distinct probabilistic and causal perspectives for understanding the phenomenon in practical application. Our research investigates data from observational studies and clinical trials on antibiotic prophylaxis for prevention of nosocomial infections -- one of the most common killers in hospitals worldwide.
Honors & Awards
-
USA Math Olympiad Training Team
Mathematical Association of America
Member of the USA Math Olympiad Training Team (MOP). The top individuals nationally are selected to represent the USA in international math competitions.
MOSP (Nat. Top 50), 2-Time USA Math Olympiad (USAMO) Finalist (Nat. Top 250, <0.01%), 5-Time AIME Qualifier, AMC12/AMC10 Distinguished Honor Rolls & Awards
☛ http://www.maa.org/math-competitions/invitational-competitions -
USA Physics Olympiad Silver Medalist
American Institute of Physics
USA Physics Olympiad Silver Medalist (Nat. Top 50), USA Physics Olympiad Honorable Mention, 2-Time USAPhO Qualifier
☛ https://www.aapt.org/physicsteam/2014/
Languages
-
English
Native or bilingual proficiency
-
Chinese
Native or bilingual proficiency
-
Spanish
Professional working proficiency
More activity by Nancy
-
We believe in merit at Moonhub 🚀, but it’s also not an excuse to be lazy. Regardless of how you evaluate candidates — all companies can benefit from…
We believe in merit at Moonhub 🚀, but it’s also not an excuse to be lazy. Regardless of how you evaluate candidates — all companies can benefit from…
Shared by Nancy Xu
-
"It's too risky joining a startup." The 30th engineer at Airbnb made $25,170,000... The opportunity to achieve greatness and build from the…
"It's too risky joining a startup." The 30th engineer at Airbnb made $25,170,000... The opportunity to achieve greatness and build from the…
Liked by Nancy Xu
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Nancy Xu in United States
-
Nancy Xu
-
Nancy Xu
MBA Candidate at Yale SOM
-
Nancy Xu
Physician (Internal Medicine, Nephrologist)
-
Nancy Xu
73 others named Nancy Xu in United States are on LinkedIn
See others named Nancy Xu