Michael Rizvi-Martel
Hey I'm Michael!
Click here for pictures of my cat
About me
Welcome to my page!
My name is Michael and I am a PhD student in Computer Science at Mila and Université de Montréal.
I am co-supervised by prof. Guillaume Rabusseau and prof. Adam Oberman.
My research interests are all over the place, but I am generally interested in the intersection between Deep Learning (DL) and Theoretical Computer Science (TCS).
On one hand, I believe that TCS presents a useful set of tools for analyzing the ability to "reason step by step" in modern language models.
On the other hand, I believe compound AI systems that use generative models and formal verification in tandem offer a great near-term solution to the development of robust AI.
Asides from this, I am equally working on projects relating to quantum circuit generation using generative AI and disentangled representation learning in SSL.
Overall, my goal is to better understand the limitations of modern AI when it comes to complex reasoning tasks and (hopefully!)
design future AI systems which overcome such limitations.
When I am not doing research, you can find me rock climbing, hiking or canoe camping!
I am also an avid guitar player and love discovering new (and old) restaurants in the Montréal area :)
News
- Our paper with Maude Lizaire was chosen for an ICML spotlight!
Publications
- Lizaire, M., Rizvi-Martel, M., Gamal, M., & Rabusseau, G. A Tensor Decomposition Perspective on Second-order RNNs. In Forty-first International Conference on Machine Learning. (Spotlight!)
- Rizvi-Martel M., Lizaire, M., Lacroce, C., & Rabusseau, G. (2024, April). Simulating weighted automata over sequences and trees with transformers. In International Conference on Artificial Intelligence and Statistics (pp. 2368-2376). PMLR.
Talks
CV
Download my CV here.
Socials
