Michel Ma

Joined

October 2021

Names

Michel Ma (Preferred)

Emails

****@mila.quebec (Confirmed)

Personal Links

Career & Education History

PhD student

University of Montreal (umontreal.ca)

2021 – 2024

MS student

University of Montreal (umontreal.ca)

2019 – 2021

Undergrad student

McGill University (mcgill.ca)

2016 – 2019

Advisors, Relations & Conflicts

No relations added

Expertise

Reinforcement learning

2019 – 2024

Publications

Automated Reward Design for Gran Turismo
Michel Ma, Takuma Seno, Kaushik Subramanian, Peter R. Wurman, Peter Stone, Craig Sherstan
- LAW
- Readers: Everyone
Do Transformer World Models Give Better Policy Gradients?
Michel Ma, Tianwei Ni, Clement Gehring, Pierluca D'Oro, Pierre-Luc Bacon
- ICML 2024 Poster
- Readers: Everyone
When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Tianwei Ni, Michel Ma, Benjamin Eysenbach, Pierre-Luc Bacon
- FMDM@NeurIPS2023
- Readers: Everyone
A Differentiable Sequence Model Perspective on Policy Gradients
Michel Ma, Pierluca D'Oro, Tianwei Ni, Clement Gehring, Pierre-Luc Bacon
- Submitted to ICLR 2024
- Readers: Everyone
Bridging State and History Representations: Understanding Self-Predictive RL
Tianwei Ni, Benjamin Eysenbach, Erfan SeyedSalehi, Michel Ma, Clement Gehring, Aditya Mahajan, Pierre-Luc Bacon
- ICLR 2024 poster
- Readers: Everyone
When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Tianwei Ni, Michel Ma, Benjamin Eysenbach, Pierre-Luc Bacon
- NeurIPS 2023 oral
- Readers: Everyone
Long-Term Credit Assignment via Model-based Temporal Shortcuts
Michel Ma, Pierluca D'Oro, Yoshua Bengio, Pierre-Luc Bacon
- 12 Oct 2021 (modified: 05 May 2023)
- Deep RL Workshop NeurIPS 2021
- Readers: Everyone

Co-Authors