“I have been working very closely with Anu for the last several years, and she is a great team member I would recommend to anyone. She is knowledgeable, detail oriented, customer focused, and I really enjoy working with her who has positive attitude. We have been working on a difficult project for a demanding Korean customer, and she managed the project very successfully, which could have turned into a nightmare. I believe she will be a great asset to any team that she will join.”
About
Activity
-
It’s a good day for everyone involved in Inflection AI. I am excited to work with incoming CEO Sean White, who has been a friend and colleague for…
It’s a good day for everyone involved in Inflection AI. I am excited to work with incoming CEO Sean White, who has been a friend and colleague for…
Liked by Anu Venkatesh
-
In the role eliminations that were effected today in parts of Amazon, some excellent personnel from my team and the broader organization were…
In the role eliminations that were effected today in parts of Amazon, some excellent personnel from my team and the broader organization were…
Liked by Anu Venkatesh
-
I had such a blast sharing the Storia AI story (and a few of my stand-up jokes) at the Amazon Web Services (AWS) Generative AI Accelerator showcase.…
I had such a blast sharing the Storia AI story (and a few of my stand-up jokes) at the Amazon Web Services (AWS) Generative AI Accelerator showcase.…
Liked by Anu Venkatesh
Experience & Education
Publications
-
Alexa Prize — State of the Art in Conversational AI
AAAI AI Magazine
To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5 million dollar competition that challenges university teams to build conversational agents, or "socialbots", that can converse coherently and engagingly with humans on popular topics for 20 minutes. The Alexa Prize offers the academic community a unique opportunity to perform research at scale with real conversational data obtained by interacting with millions of Alexa users, along with user-provided…
To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5 million dollar competition that challenges university teams to build conversational agents, or "socialbots", that can converse coherently and engagingly with humans on popular topics for 20 minutes. The Alexa Prize offers the academic community a unique opportunity to perform research at scale with real conversational data obtained by interacting with millions of Alexa users, along with user-provided ratings and feedback, over several months. This enables teams to effectively iterate, improve and evaluate their socialbots throughout the competition. Sixteen teams were selected for the inaugural competition last year. To build their socialbots, the students combined state-of-the-art techniques with their own novel strategies in the areas of Natural Language Understanding and Conversational AI. This article reports on the research conducted over the 2017-2018 year. While the 20 minute grand challenge was not achieved in the first year, the competition produced several conversational agents that advanced the state of the art, are interesting for everyday users to interact with, and help form a baseline for the second year of the competition.
-
Contextual Language Model Adaptation for Conversational Agents
Interspeech
Statistical language models (LM) play a key role in Automatic Speech Recognition (ASR) systems used by conversational agents. These ASR systems should provide a high accuracy under a variety of speaking styles, domains, vocabulary and argots. In this paper, we present a DNN-based method to adapt the LM to each user-agent interaction based on generalized contextual information, by predicting an optimal, context-dependent set of LM interpolation weights. We show that this framework for contextual…
Statistical language models (LM) play a key role in Automatic Speech Recognition (ASR) systems used by conversational agents. These ASR systems should provide a high accuracy under a variety of speaking styles, domains, vocabulary and argots. In this paper, we present a DNN-based method to adapt the LM to each user-agent interaction based on generalized contextual information, by predicting an optimal, context-dependent set of LM interpolation weights. We show that this framework for contextual adaptation provides accuracy improvements under different possible mixture LM partitions that are relevant for both (1) Goal-oriented conversational agents where it's natural to partition the data by the requested application and for (2) Non-goal oriented conversational agents where the data can be partitioned using topic labels that come from predictions of a topic classifier. We obtain a relative WER improvement of 3% with a 1-pass decoding strategy and 6% in a 2-pass decoding framework, over an unadapted model. We also show up to a 15% relative improvement in recognizing named entities which is of significant value for conversational ASR systems.
-
Conversational AI: The science behind the Alexa Prize
Alexa Prize Proceedings
Conversational agents are exploding in popularity. However, much work remains in the area of social conversation as well as free-form conversation over a broad range of domains and topics. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million-dollar university competition where sixteen selected university teams were challenged to build conversational agents, known as socialbots, to converse coherently and engagingly with humans on popular topics…
Conversational agents are exploding in popularity. However, much work remains in the area of social conversation as well as free-form conversation over a broad range of domains and topics. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million-dollar university competition where sixteen selected university teams were challenged to build conversational agents, known as socialbots, to converse coherently and engagingly with humans on popular topics such as Sports, Politics, Entertainment, Fashion and Technology for 20 minutes. The Alexa Prize offers the academic community a unique opportunity to perform research with a live system used by millions of users. The competition provided university teams with real user conversational data at scale, along with the user-provided ratings and feedback augmented with annotations by the Alexa team. This enabled teams to effectively iterate and make improvements throughout the competition while being evaluated in real-time through live user interactions. To build their socialbots, university teams combined state-of-the-art techniques with novel strategies in the areas of Natural Language Understanding, Context Modeling, Dialog Management, Response Generation, and Knowledge Acquisition. To support the efforts of participating teams, the Alexa Prize team made significant scientific and engineering investments to build and improve Conversational Speech Recognition, Topic Tracking, Dialog Evaluation, Voice User Experience, and tools for traffic management and scalability. This paper outlines the advances created by the university teams as well as the Alexa Prize team to achieve the common goal of solving the problem of Conversational AI.
-
On Evaluating and Comparing Conversational Agents
NIPS 2017 Conversational AI Workshop
Conversational agents are exploding in popularity. However, much work remains in the area of non goal-oriented conversations, despite significant growth in research interest over recent years. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million dollar university competition where sixteen selected university teams built conversational agents to deliver the best social conversational experience. Alexa Prize provided the academic community with the…
Conversational agents are exploding in popularity. However, much work remains in the area of non goal-oriented conversations, despite significant growth in research interest over recent years. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million dollar university competition where sixteen selected university teams built conversational agents to deliver the best social conversational experience. Alexa Prize provided the academic community with the unique opportunity to perform research with a live system used by millions of users. The subjectivity associated with evaluating conversations is key element underlying the challenge of building non-goal oriented dialogue systems. In this paper, we propose a comprehensive evaluation strategy with multiple metrics designed to reduce subjectivity by selecting metrics which correlate well with human judgement. The proposed metrics provide granular analysis of the conversational agents, which is not captured in human ratings. We show that these metrics can be used as a reasonable proxy for human judgment. We provide a mechanism to unify the metrics for selecting the top performing agents, which has also been applied throughout the Alexa Prize competition. To our knowledge, to date it is the largest setting for evaluating agents with millions of conversations and hundreds of thousands of ratings from users. We believe that this work is a step towards an automatic evaluation process for conversational AIs.
-
Topic-based evaluation for conversational bots
NIPS 2017 Conversational AI workshop paper
Dialog evaluation is a challenging problem, especially for non task-oriented dialogs where conversational success is not well-defined. We propose to evaluate dialog quality using topic-based metrics that describe the ability of a conversational bot to sustain coherent and engaging conversations on a topic, and the diversity of topics that a bot can handle. To detect conversation topics per utterance, we adopt Deep Average Networks (DAN) and train a topic classifier on a variety of question and…
Dialog evaluation is a challenging problem, especially for non task-oriented dialogs where conversational success is not well-defined. We propose to evaluate dialog quality using topic-based metrics that describe the ability of a conversational bot to sustain coherent and engaging conversations on a topic, and the diversity of topics that a bot can handle. To detect conversation topics per utterance, we adopt Deep Average Networks (DAN) and train a topic classifier on a variety of question and query data categorized into multiple topics. We propose a novel extension to DAN by adding a topic-word attention table that allows the system to jointly capture topic keywords in an utterance and perform topic classification. We compare our proposed topic based metrics with the ratings provided by users and show that our metrics both correlate with and complement human judgment. Our analysis is performed on tens of thousands of real human-bot dialogs from the Alexa Prize competition and highlights user expectations for conversational bots.
-
Healthbuzz: Contextual Social search and Conversations
Search in Social Media
Healthbuzz is a health specific social search and conversation system being developed for Georgia Tech (GT) student community. It allows users to search health issues or start a new health specific conversation and provides three types of contextual results: related conversations to join, community people to talk to with similar profiles, and Web search results based on users’ search criteria or on-going conversation. We found out that people connections were a crucial aspect of the Healthbuzz…
Healthbuzz is a health specific social search and conversation system being developed for Georgia Tech (GT) student community. It allows users to search health issues or start a new health specific conversation and provides three types of contextual results: related conversations to join, community people to talk to with similar profiles, and Web search results based on users’ search criteria or on-going conversation. We found out that people connections were a crucial aspect of the Healthbuzz system that engaged the partici- pants and provided greater user satisfaction. To our surprise, users did not show an active interest in the web based recommendations during a conversation, unless pointed to explicitly by one of the participants. We briefly present the system and discuss the findings of the usability study of the system.
Other authorsSee publication -
Cobot: Real Time Multi User Conversational Search and Recommendations
Recommender Systems and The Social Web at ACM RecSys
Cobot is a new intelligent agent platform that connects users through real-time and off-line conversations about their health and medical issues. Intelligent web based information agents (conversational/community bots) participate in each conver- sation providing highly-relevant real-time informational rec- ommendations and connecting people with relevant conver- sations and other community members. Cobot provides an innovative approach to facilitate easier information access al- lowing users…
Cobot is a new intelligent agent platform that connects users through real-time and off-line conversations about their health and medical issues. Intelligent web based information agents (conversational/community bots) participate in each conver- sation providing highly-relevant real-time informational rec- ommendations and connecting people with relevant conver- sations and other community members. Cobot provides an innovative approach to facilitate easier information access al- lowing users to exchange information through a natural lan- guage conversational approach. Conversational Search(CS) is an interactive and collaborative information finding in- teraction. The participants in this interaction engage in social conversations aided with an intelligent information agent (Cobot) that provides contextually relevant factual, web search and social search recommendations. Cobot aims to help users make faster and more informed search and dis- covery. It also helps the agent learn about conversations with interactions and social feedback to make better recom- mendations. Cobot leverages the social discovery process by integrating web information retrieval along with the social interactions and recommendations.
Other authorsSee publication -
Collaborative Information Access: A Conversational Search Approach
WebCBR at International Conference on Case Based Reasoning
Knowledge and user generated content is proliferating on the web in scientific publications, information portals and online social me- dia. This knowledge explosion has continued to outpace technological innovation in efficient information access technologies. In this paper, we describe the methods and technologies for ‘Conversational Search’ as an innovative solution to facilitate easier information access and reduce the information overload for users. Conversational Search is an interactive…
Knowledge and user generated content is proliferating on the web in scientific publications, information portals and online social me- dia. This knowledge explosion has continued to outpace technological innovation in efficient information access technologies. In this paper, we describe the methods and technologies for ‘Conversational Search’ as an innovative solution to facilitate easier information access and reduce the information overload for users. Conversational Search is an interactive and collaborative information finding interaction. The participants in this interaction engage in social conversations aided with an intelligent information agent (Cobot) that provides contextually relevant search recommendations. The collaborative and conversational search activity helps users make faster and more informed search and discovery. It also helps the agent learn about conversations with interactions and social feedback to make better recommendations. Conversational search lever- agesthe social discovery process by integrating web information retrieval along with the social interactions.
Other authorsSee publication -
iReMedI – Intelligent Retrieval of Medical Information
European Conference on Case Based Reasoning
Effective encoding of information is one of the keys to qual- itative problem solving. Our aim is to explore Knowledge representation techniques that capture meaningful word associations occurring in doc- uments. We have developed iReMedI, a TCBR based problem solving system as a prototype to demonstrate our idea. For representation we have used a combination of NLP and graph based techniques which we call as Shallow Syntactic Triples, Dependency Parses and Semantic Word Chains. To test their…
Effective encoding of information is one of the keys to qual- itative problem solving. Our aim is to explore Knowledge representation techniques that capture meaningful word associations occurring in doc- uments. We have developed iReMedI, a TCBR based problem solving system as a prototype to demonstrate our idea. For representation we have used a combination of NLP and graph based techniques which we call as Shallow Syntactic Triples, Dependency Parses and Semantic Word Chains. To test their effectiveness we have developed retrieval techniques based on PageRank, Shortest Distance and Spreading Activation meth- ods. The various algorithms discussed in the paper and the comparative analysis of their results provides us with useful insight for creating an effective problem solving and reasoning system.
Other authorsSee publication
Courses
-
Advanced Internet Application Development
-
-
Case Based Reasoning
-
-
Computability and Algorithms
-
-
Computer Science Ventures
-
-
Data Structures
-
-
Decision Support Systems
-
-
Digital Image Processing
-
-
Entrepreneurship Forum
-
-
High Performance Computer Architecture
-
-
Knowledge Based AI
-
-
Machine Learning
-
-
Natural Language Processing
-
-
OOP with C++
-
-
Object Oriented Analysis and Design
-
-
Operating Systems
-
Languages
-
English
-
-
Hindi
-
-
Kannada
-
Recommendations received
3 people have recommended Anu
Join now to viewMore activity by Anu
-
I'm excited to be speaking at Tech Summit SF on Thursday (6/22). I'll be talking about Alexa's ability to process data from multiple input sensors…
I'm excited to be speaking at Tech Summit SF on Thursday (6/22). I'll be talking about Alexa's ability to process data from multiple input sensors…
Liked by Anu Venkatesh
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 Anu Venkatesh in United States
-
Anu Venkatesh
Incoming Applied Math + English at UC San Diego
-
Anu Venkatesh
-
Anu Venkatesh, MBA, MA, AMP
-
Anu VENKATESH
--
4 others named Anu Venkatesh in United States are on LinkedIn
See others named Anu Venkatesh