About
I am a Postdoctoral Fellow in the Montreal Robotics and Embodied AI Laboratory (REAL) at the Université de Montréal and the Mila-Quebec Artificial Intelligence Institute, where I work with Liam Paull. I am interested in several areas of robotic planning, perception, and learning, but a common theme in my research is the assertion that robots are complex tools that must interface with humans at many stages of design, development, and deployment in order to achieve their full effect. My work often focuses on how users and developers can be factored in to both the design and operation of the robot to improve both autonomy and overall system function, in a sense re-designing different robotics subsystems to leverage reliable human resources within pre-, on-, and post-deployment frameworks. Most recently I have been working on several reinforcement learning projects investigating how and why different forms of state augmentation and constraint specification (some from human data) can lead to safer, more efficient policies and policy learning.
Before joining REAL, I completed my Ph.D. in computer science at the University of Massachusetts Amherst, where I found a home in both the Laboratory for Perceptual Robotics and the Resource Bounded Reasoning Laboratory, directed by Rod Grupen and Shlomo Zilberstein, respectively. My thesis proposed Multi-SLAM systems, a meta-reasoning framework for dynamically selecting between multiple SLAM front-end modules depending on the sensor stimuli afforded by the current environment. For example, in low-light lasers or structured light may be better, and in clutter or bright light cameras may be more reliable. Prior to graduate school, I earned a B.A. in physics and computer science from Swarthmore College.
News
Jan. 2024: Our paper proposing Multi-SLAM systems was accepted to ICRA 2024!
Jan. 2024: Our paper on Ethically Compliant Sequential Decision Making under Partial Observability was accepted to ICRA 2024!
Jan. 2024: I was awarded a US patent along with my collaborators at Amazon on physical models for hierarchical clustering and symbolic inference!
Nov. 2023: Our paper on Q-value Augmentation for Meta-Reinforcement Learning was accepted to the GenPlan workshop at Neurips and will be highlighted as a talk!
Nov. 2023: I will be joining REAL as a posdoctural fellow!
Nov. 2023: I defended my thesis!
July 2023: I gave a talk on Achieving Robust SLAM Systems via Integration throughout the Robotics Life Cycle at REAL
July 2023: I gave a talk on the Limits, Lessons, and Opportunities of Fairness and Sequential Decision Making at the EQUAL Lab.
July 2023: Our paper on Formal Composition of Robotic Systems as Contract Programs was accepted to IROS 2023!
June 2023: I presented recent work on Causal Explanations for Sequential Decision Making under Uncertainty at the 7th Annual Center for Human-Compatible AI Workshop.
Mar. 2023: I successfully defended my thesis proposal! More work to come on Multi-SLAM systems for fault tolerant localization and mapping!
Jan. 2023: I’ve been awarded a Dissertation Writing Fellowship from the College of Information and Computer Sciences at UMass!
Nov. 2022: I was awarded a US patent along with my collaborators at Amazon Robotics on motion planning in dynamic enironments!
July 2022: Our paper on learning to form partial abstractions of MDPs was accepted to IROS 2022!
Apr. 2022: I gave a talk on Generating Explanans for Markov Decision Processes at the General Motors Forum on Explainable Artificial Intelligence.
Mar. 2022: One of my MS student mentees, Chhandak Bagchi, has been accepted into the PhD program at UMass Amherst!
Feb. 2022: Our paper on learning to tune parameters in anytime algorithms was accepted to ICAPS 2022!
Oct. 2021: Our workshop at IROS 2021 on Building and Evaluating Ethical Robotic Systems was a success! Thank you to all of our wonderful speakers and contributors!
Sep. 2021: I was awarded a US patent along with my collaborators at Nissan on localization for autonomous vehicles!
July 2021: Our survey on opponent modeling was accepted for publication in the Journal of Artificial Intelligence Research
July 2021: Our paper on Robust Rank-Deficient SLAM was accepted to IROS 2021!
May 2021: Our workshop on Building and Evaluating Ethical Robotic Systems was accepted and will be held at IROS 2021!
Mar. 2021: Our paper on Ethically Compliant Planning in Moral Communities was accepted to AIES 2021!
Feb. 2021: Our paper on Ethically Compliant Sequential Decision Making was recognized by a Distinguished Paper Award at AAAI!
Jan. 2021: Our paper on Solving MDPs using Partial Abstractions was accepted to ICRA 2021!
Dec. 2020: Our paper on Ethically Compliant Sequential Decision Making was accepted to AAAI 2021!
Nov. 2020: My survey on loop closure detection and its potential as a good task for evaluating intelligent agents was accepted to the Workshop on Machine Learning Retrospectives, Surveys & Meta-Analyses at NeurIPS 2020!
July 2020: My paper on learning embedded representations for efficient storage and query support for robotic perception data was accepted to IROS 2020!
May 2020: I’ve been awarded a Massachusetts Space Grant Consortium Fellowship!
Aug. 2019: I will be staying on at Amazon Lab126 through the fall. Very exited to continue my project!
June 2019: I’ve offically received my Masters degree in computer science!
May 2019: I’ve passed my portfolio (qualifications) and I am officially a PhD Candidate!
Feb. 2019: I’ll be headed to Sunnyvale California this summer to work at Amazon Lab126!
Feb. 2018: I’ll be headed to North Reading Massachusetts this summer to work at Amazon Robotics!
Jan. 2018: Our paper on lane localization for autonomous vehicles was accepted to ICRA 2018!
Dec. 2017: Our paper on Human-in-the-Loop SLAM has been accepted to AAAI 2018!
Feb. 2017: I’ll be headed to Sunnyvale California this summer to work at Nissan Research Center on their autonomous vehicle program!
July 2016: Our paper on long-term vector mapping was accepted to IROS 2016!