I am a postdoctoral researcher at Microsoft Research (MSR). Within Microsoft Research, I am part of the Machine Learning research area, with personal focus on Reinforcement Learning and sequential decision making problems more generally.
I previously defended my PhD in Computing + Mathematical Sciences at Caltech. I was fortunate to be advised by Professor Yisong Yue. My other committee members include Anima Anandkumar Yue, Adam Wierman, and Hal Daume. My thesis focuses on systematic approaches to combine domain knowledge and learning methods to solve sequential decision making problems, which span reinforcement learning, imitation learning, optimal control. My recent works have been applied to diverse set of domains including robotics, autonomous camera planning for broadcasting, modeling human decision making in team sports, video games and simulated car racing. During grad school years I spent two summers interning at Microsoft Research in NYC and Disney Research in Pittsburgh. I was also supported by an Amazon AI Graduate fellowship.
- I defended my PhD in Computing + Mathematical Sciences from Caltech in fall 2019. I had lots of fun formulating new research directions in Structured Policy Learning.
- I organized a workshop titled Real-World Sequential Decision Making: Reinforcement Learning and Beyond at the International Conference on Machine Learning (ICML) in summer 2019.
- I gave an invited talk at the 2019 American Control Conference workshop on the interplay between Control, Optimization, and Machine Learning in summer 2019.
- I gave a tutorial on Imitation Learning, together with Yisong Yue at the International Conference on Machine Learning (ICML) in Stockholm in July 2018
Preprints
Cameron Voloshin, Hoang M. Le, Nan Jiang and Yisong Yue
Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning
(under review) , 2019
[PDF] [BibTeX]@article{voloshin2019empirical, title={Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning}, author={Voloshin, Cameron and Le, Hoang M and Jiang, Nan and Yue, Yisong}, journal={arXiv preprint arXiv:1911.06854}, year={2020} }
Recent Publications
Dimitar Ho , Hoang M. Le, John Doyle, and Yisong Yue
Online Robust Control of Nonlinear Systems with Large Uncertainty
International Conference on Artificial Intelligence and Statistics (AISTATS) , 2021 [PDF] [BibTeX]@inproceedings{ho2021online, title={Online Robust Control of Nonlinear Systems with Large Uncertainty}, author={Ho, Dimitar and Le, Hoang and Doyle, John and Yue, Yisong}, booktitle={International Conference on Artificial Intelligence and Statistics}, pages={3475--3483}, year={2021}, organization={PMLR} }Minh Pham , Anh Ninh, Hoang M. Le, and Yufeng Liu
An Efficient Algorithm for Minimizing Multi Non-Smooth Component Functions
Journal of Computational and Graphical Statistics , 2020 [PDF] [BibTeX]@article{pham2020efficient, title={An efficient algorithm for minimizing multi non-smooth component functions}, author={Pham, Minh and Ninh, Anh and Le, Hoang and Liu, Yufeng}, journal={Journal of Computational and Graphical Statistics}, pages={1--9}, year={2020}, publisher={Taylor \& Francis} }Abhinav Verma* , Hoang M. Le*, Yisong Yue and Swarat Chaudhuri
Imitation-Projected Programmatic Reinforcement Learning
Neural Information Processing Systems (NeurIPS) , 2019
(*equal contribution)
[PDF] [Summary slides] [BibTeX]@article{verma2019imitation, title={Imitation-Projected Programmatic Reinforcement Learning}, author={Verma, Abhinav and Le, Hoang M and Yue, Yisong and Chaudhuri, Swarat}, journal={arXiv preprint arXiv:1907.05431}, year={2019} }Hoang M. Le, Cameron Voloshin and Yisong Yue
Batch Policy Learning under Constraints
International Conference on Machine Learning (ICML), 2019
(Long Talk)
[PDF] [Project Page] [BibTeX]@article{le2019batch, title={Batch Policy Learning under Constraints}, author={Le, Hoang and Voloshin, Cameron and Yue, Yisong}, booktitle={International Conference on Machine Learning}, pages={3703--3712}, year={2019} }Andrew J. Taylor, Victor D. Dorobantu, Hoang M. Le, Yisong Yue and Aaron D. Ames
Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems
International Conference on Intelligent Robots and Systems (IROS) , 2019
[PDF][code] [BibTeX]@article{taylor2019episodic, title={Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems}, author={Taylor, Andrew J and Dorobantu, Victor D and Le, Hoang M and Yue, Yisong and Ames, Aaron D}, journal={arXiv preprint arXiv:1903.01577}, year={2019} }Andrew J. Taylor, Victor D. Dorobantu, Meera Krisnamoorthy, Hoang M. Le, Yisong Yue and Aaron D. Ames
A Control Lyapunov Perspective on Episodic Learning via Projection to State Stability
IEEE Conference on Decision and Control (CDC) , 2019
[PDF] [BibTeX]@article{taylor2019acontrol, title={A Control Lyapunov Perspective on Episodic Learning via Projection to State Stability}, author={Taylor, Andrew J and Dorobantu, Victor D and Krisnamoorthy, Meera and Le, Hoang M and Yue, Yisong and Ames, Aaron D}, journal={arXiv preprint arXiv:1903.07214}, year={2019} }Hoang M. Le, Nan Jiang, Alekh Agarwal, Miroslav Dudík, Yisong Yue and Hal Daumé III
Hierarchical Imitation and Reinforcement Learning
International Conference on Machine Learning (ICML), 2018
(Long Talk)
[PDF] [Project Page] [BibTeX]@inproceedings{le2018hierarchical, title={Hierarchical Imitation and Reinforcement Learning}, author={Le, Hoang and Jiang, Nan and Agarwal, Alekh and Dudik, Miroslav and Yue, Yisong and Daum{\'e}, Hal}, booktitle={International Conference on Machine Learning}, pages={2923--2932}, year={2018} }Hoang M. Le, Yisong Yue, Peter Carr and Patrick Lucey
Coordinated Multi-Agent Imitation Learning
International Conference on Machine Learning (ICML), 2017
[PDF] [long] [BibTeX]@inproceedings{le2017coordinated, title={Coordinated Multi-Agent Imitation Learning}, author={Le, Hoang M and Yue, Yisong and Carr, Peter and Lucey, Patrick}, booktitle={International Conference on Machine Learning}, pages={1995--2003}, year={2017} }Hoang M. Le, Peter Carr,Yisong Yue and Patrick Lucey
Data-Driven Ghosting using Deep Imitation Learning
MIT Sloan Sports Analytics Conference (SSAC), 2017
(Best Paper Runner-Up Award)
[PDF] [Project Page] [Video Demo] [BibTeX]@article{le2017data, title={Data-Driven Ghosting using Deep Imitation Learning}, author={Le, Hoang M and Carr, Peter and Yue, Yisong and Lucey, Patrick} journal={MIT Sloan Sports Analytics Conference (SSAC)}, year={2017} }Hoang M. Le, Andrew Kang, Yisong Yue and Peter Carr
Smooth Imitation Learning for Online Sequence Prediction
International Conference on Machine Learning (ICML), 2016
[PDF] [long] [BibTeX]@inproceedings{le2016smooth, title={Smooth Imitation Learning for Online Sequence Prediction}, author={Le, Hoang and Kang, Andrew and Yue, Yisong and Carr, Peter}, booktitle={International Conference on Machine Learning}, pages={680--688}, year={2016} }Jianhui Chen, Hoang M. Le, Peter Carr, Yisong Yue and James J. Little
Learning Online Smooth Predictors for Real-time Camera Planning using Recurrent Decision Trees
Computer Vision and Pattern Recognition (CVPR), 2016
(Oral Presentation)
[PDF] [Press Release] [BibTeX]@inproceedings{chen2016learning, title={Learning online smooth predictors for realtime camera planning using recurrent decision trees}, author={Chen, Jianhui and Le, Hoang M and Carr, Peter and Yue, Yisong and Little, James J}, booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition}, pages={4688--4696}, year={2016} }
Pending Patents
George Peter Kenneth Carr, Hoang M. Le, Yisong Yue
Data-Driven Ghosting Using Deep Imitation Learning
[patent pending]