Publications

What can I do here? A Theory of Affordances in Reinforcement Learning [Paper Coming Soon]

Khimya Khetarpal, Zafarali Ahmed, Gheorghe Comanici, David Abel, Doina Precup. (ICML)(2020) *Work done during an internship at DeepMind

Options of Interest: Temporal Abstraction with Interest Functions [Paper] [Poster] [Project Webpage]

Khimya Khetarpal, Martin Klissarov, Maxime Chevalier-Boisvert, Pierre-Luc Bacon, Doina Precup. Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI), 2020.

previously presented at NeurIPS 2019 Workshop on Deep Reinforcement Learning, NeurIPS 2019 Workshop on Learning Transferable Skills.

Value Preserving State Action Abstractions [Paper]

David Abel, Nathan Umbanhowar, Khimya Khetarpal, Dilip Arumugam, Doina Precup, Michael L. Littman. *To appear International Conference on Artificial Intelligence and Statistics (AISTATS), 2020.

previously presented at ICLR Workshop on Structure and Priors in RL(2019), The Multi-disciplinary Conference on Reinforcement Learning & Decision Making, RLDM (2019)

Variational State Encoding as Intrinsic Motivation in Reinforcement Learning

Martin Klissarov, Riashat Islam, Khimya Khetarpal, Doina Precup. Task Agnostic Reinforcement Learning Workshop ICLR (2019), The Multi-disciplinary Conference on Reinforcement Learning & Decision Making, RLDM (2019)

Learning Generalized Temporal Abstractions Across Both Action and Perception [Paper]

Khimya Khetarpal, Doina Precup. AAAI/SIGAI Doctoral Consortium (DC) at AAAI (2019) *Awarded Scholarship Funding

Learning Options with Interest Functions [Paper] [Poster]

Khimya Khetarpal, Doina Precup AAAI Student Abstract (SA) at AAAI (2019) *Finalist for 3MT Competition

Environments for Lifelong Reinforcement Learning [Paper] [Poster]

Khimya Khetarpal*, Shagun Sodhani*, Sarath Chandar, Doina Precup. Continual Learning Workshop, NeurIPS 2018

Attend Before you Act: Leveraging Human Visual Attention for Continual Learning [Project Page] [Paper] [Poster] [Code]

Khimya Khetarpal, Doina Precup. Lifelong Learning: A Reinforcement Learning Approach (LLARLA) Workshop, ICML 2018 Spotlight, Oral presentation. *Best Student Paper Award

Safe Option-Critic: Learning Safety in the Option-Critic Architecture [Project Page] [Paper] [Code]

Arushi Jain, Khimya Khetarpal, Doina Precup. Adaptive Learning Agents (ALA) Workshop, ICML 2018

RE-EVALUATE: Reproducibility in Evaluating Reinforcement Learning Algorithms [Paper] [Poster] [Code]

Khimya Khetarpal*, Zafarali Ahmed*, Andre Cianflone, Riashat Islam, Joelle Pineau. Reproducibility in Machine Learning Workshop, ICML 2018

Creating Segments and Effects on Comics by Clustering Gaze Data [Project Page] [Paper] [Code]

I. Thirunarayanan, Khimya Khetarpal, S. J. Koppal, J. Shea, O Le Meur & E. Jain. ACM – Transactions on Multimedia Computing, Communications and Applications, TOMM 2017

A Preliminary Benchmark Of Four Saliency Algorithms On Comic Art [Project Page] [Paper]

Khimya Khetarpal, Eakta Jain. International Workshop on Multi Media Artwork Analysis, ICME 2016

Mobile robot navigation using evolving neural controller in unstructured environments [Paper] [Code] [Talk Slides]

Awhan Patnaik, Khimya Khetarpal, Laxmidhar Behera. International Conference on – Advances in Control and Optimization of Dynamical Systems, ACODS 2014

Advanced Adaptive Safety Control System [Paper]

Khimya Khetarpal, Kshitij Gupta, B. Karthikeyan, C. Sasi Kumar. International Conference on – Informatics in Control Automation & Robotics, ICINCO 2012

17/7/2020 Our research on "What can I do here? A Theory of Affordances in RL" is featured at the MIT Technology Review.

1/6/2020 Our paper on "What can I do here? A Theory of Affordances in RL" has been accepted at the International Conference on Machine Learning (ICML) 2020. See full paper here.

14/5/2020 I gave a talk on A Theory of Affordances in RL at the Reinforcement Learning and Artificial Intelligence lab meeting at University of Alberta, invited by Martha White.

6/1/2020 Our work on Value Preserving State Action Abstractions has been accepted for presentation at AISTATS, 2020.

11/10/2019 Our work on Options of Interest: Temporal Abstraction with Interest Functions has been accepted at AAAI, 2020. A preliminary draft of the ideas in this work was published in the AAAI, 2019 Student Abstract proceedings

13/9/2019 I taught an invited lecture on “Introduction to RL” at the MILA/IVADO summer school on Deep Learning, Montreal, 2019

29/7/2019 I am currently a PhD Research Intern at DeepMind , Montreal working with Doina Precup and the HRL team.

6/6/2019 I am co-organizing a Lifelong Learning: A Reinforcement Learning (LLARLA) Workshop to be held during RLDM, 2019 .

24/3/2019 I am co-organizing a Workshop on Multi-Task and Lifelong Reinforcement Learning to be held during ICML-19 in Long Beach, USA . We have an exicting line up of speakers. Consider submitting your work.

10/26/2018 Our work on Learning Options with Interest Functions, has been selected for poster presentation in the AAAI-19 Student Abstract and Poster Program to be held during AAAI-19 in Honolulu, Hawaii

10/15/2018 I have been chosen to be a presenter at the 24th AAAI/SIGAI Doctoral Consortium (DC) at AAAI-19 in Honolulu, Hawaii

6/20/2018 Our work on 'Safe Option-Critic: Learning Safety in the Option-Critic Architecture' Accepted in ICML 2018 Workshop on Adaptive Learning Agents. Paper is now on arxiv!

6/16/2018 Our work on 'Attend Before you Act: Leveraging human visual attention for continual learning' Accepted in ICML 2018 Workshop on LifeLong Learning. Paper coming out soon!

5/13/2018 I will be one of the Teaching Assistants at the AI For Social Good Summer School from May 14-June 21. Super excited to be a part of this initiative.

2/28/2018 Invited talk on Introduction to Computer Vision & Reinforcement Learning at ITTT (Informative Talks on Technical Topics) organized by the McGill IEEE Student Branch.

11/13/2017 Volunteered as a mentor at the Women in Innovation and Artificial Intelligence Breakfast organized as the kick-off event of the McGill Innovation Week in the Fall of 2017.

10/1/2017 Organized the weekly Fall-17 AI Safety RL Lab Reading Group to get started with discussions and initial reading on related literature.

9/15/2017 Joined the Reasoning and Learning lab at McGill University as a 1st year Ph.D. student. Attended the ACM Canadian Women Celebration of Women in Computing

6/1/2017 Left job at Intel. I will be joining Ph.D. in CS program at the McGill University in Fall 2017.

5/1/2017 My application for Reinforcement Learning Summer School (RLSS) 2017 has been accepted. 225 out of 1130 applicants were selected.

4/14/2017 My application for International Computer Vision Summer School (ICVSS) 2017 has been accepted. 150 out of 430 applicants were selected.

03/27/2017 Manuscript entitled “Creating Segments and Effects on Comics by Clustering Gaze Data” has been accepted for publication in the Transactions on Multimedia Computing Communications and Applications [ACM TOMM]

02/27/2017 Invited Talk on Learning Visual Representations at the ASU Active Perception Group Seminar Series at the Arizona State University, Tempe.

08/03/2016 Invited Talk at the Women in Deep Learning event at the Deep Learning Summer School, Montreal

06/09/2016 As of June-27-2016, I start working as a Software Engineer in the Perception Computing Group-IOTG at Intel Corporation, Arizona.

05/11/2016 My application for CIFAR-CRM 2016 Deep Learning Summer School has been accepted.

04/22/2016 Paper on 'A Preliminary Benchmark of Four Saliency Algorithms On Comic Art' Accepted in ICME-MMArt Workshop 2016 .

09/23/2015 Attended RoboBusiness Expo 2015 at San Jose from Intel.

02/05/2015 I will be spending Summer & Fall of 2015 as a Perception Computing Software Intern at Intel.

12/10/2014 My robot AUMORO built in the course IMDL- Fall 2014 was covered in the WUFT news on Robot Demo Day .

Publications

Follow Blog via Email

Tags