group

Abdel Mfougouon Njupoun

(co-supervised by Blake Richards)

I am a PhD student in Applied Mathematics with a research focus at the intersection of Artificial Intelligence and Neuroscience. I have an interdisciplinary background in Mathematics, Robotics, and Engineering in Robotic Systems. I am passionate about understanding how the brain works and enthusiastic about using mathematics to explore complex systems. I love all kinds of sports, especially soccer. I also love watching documentaries.

Amine Natik

(co-supervised by Guy Wolf)

I’m a PhD student at the University of Montreal & Mila. I completed my MSc in Probability Theory at the University of Ottawa. In my research, I use tools from manifold learning, graph signal processing, representation learning, and dynamical systems. I am also interested in research at the intersection of AI and Neuroscience.

Eric Elmoznino

(co-supervised by Yoshua Bengio)

I work at the intersection of artificial intelligence and computational neuroscience, and I’m interested in understanding conscious cognition so that we can close the gap between deep learning and the human mind. While current deep neural networks excel in perceptual domains such as vision, they are unable to perform the kind of higher-order reasoning that in humans is associated with conscious, effortful thought. A core hypothesis motivating my work is that conscious representations are simple, discrete, and, most importantly, compositional. My main projects aim to formalize these elusive concepts, better understand how they work in the brain, and develop deep neural network models of them that are loosely inspired by cognitive science.

Ezekiel (Zeke) Williams

I am an applied math doctoral candidate working to uncover principles underlying learning, memory, and prediction, and how these phenomena arise from electrical dynamics inside the brain. My current work is focused on the role of randomness in neural computation, and how biological cell networks are able to learn even when the information to be learned is spread across long timescales. Given the similarity between models of the brain and AI, my work has some relevance to machine learning. Mathematically, most of my interests lie within dynamical systems theory, probability theory, and statistics. I also enjoy making music, the outdoors, Science for the People, and listening to Rage Against the Machine.

François Paugam

(co-supervised by Lune Bellec)

I am a PhD student in computer science. My main research interest is to explore techniques to incorporate brain imaging data in the training of deep neural networks as a way to make the resulting artificial agents more human-like in their behavior and in particular to improve their generalizability to new contexts.

Jean-Pierre Falet

(co-supervised by Yoshua Bengio)

I am a PhD student in computer science working on reasoning in AI systems. My research is inspired by our understanding of human cognition, and aims to integrate inductive biases for human-like cognition into AI systems. To achieve this, I leverage tools from probabilistic graphical modeling and GFlowNets. Additionally, I work on the problem of AI safety in highly capable systems, and am interested in AI research ethics and governance. Given that I am also a practicing neurologist with a research background in applied machine learning, I am interested in applying my theoretical contributions to clinical tasks.

Léo Choinière

(co-supervised by Numa Dancause)

I am a PhD student in neuroscience at Université de Montréal. I am interested in brain-machine interfaces, neural dynamics, and motor control. My main project is aimed at optimizing parameters in the context of neurostimulation in preclinical animal models. In general, my research is situated at the intersection of and requires skills from both experimental and computational neuroscience.

Léo Gagnon

I am interested in understanding general learning systems such as LLMs in order to contribute to making them better. Specifically, using ideas from meta-learning, Bayesian statistics and other relevant fields, I study foundational models in controlled settings in search of principles relating their pre-training data, mechanisms and architecture. I also like playing the guitar, spending calories, and asking deep questions.

Nanda H Krishna

I am a PhD student working on problems at the intersection of Computational Neuroscience and Deep Learning. I am specifically interested in developing a theoretical understanding of how learning shapes computation in brains and artificial neural networks. I also work on building foundation models for neuroscience, with the goal of improving decoding in brain-computer interfaces. In my free time, you can find me obsessing over figures, watching cricket, reading comics, or coming up with Korean phrases that annoy Avery.

Pravish Sainath

(co-supervised by Lune Bellec)

I’m a PhD student at Université de Montréal. The purpose of my life and study is to better understand the nature and mechanisms of both natural and artificial intelligence. I am mostly interested in using computational and cognitive neuroscience to improve deep and reinforcement learning methods and applying deep learning methods to better understand the brain.

Sangnie Bhardwaj

(co-supervised by Hugo Larochelle)

I’m a PhD student at UdeM and also a researcher at Google Montreal, where I work on building robust and generalizable image representations. I am interested in achieving this by incorporating principles inspired by visual learning in human brains.

Sarthak Mittal

(co-supervised by Yoshua Bengio)

I am a PhD student interested in designing unified AI systems that are not constrained to solving a single task but learn general-purpose optimization routines implicitly in their inference mechanism, allowing them to generalize to novel tasks in a few-shot manner. In search of this goal, my explorations revolve around architectural inductive biases and representation learning to encourage compositional computations in Neural Networks as well as generative modeling and Bayesian inference methodologies akin to probabilistic meta-learning setups. A core guiding principle of my work is to efficiently model some crucial invariances underlying a vast majority of tasks, for example optimization procedures, and I believe that leveraging AI systems to model exactly these invariances would lead to few-shot generalization capabilities across novel tasks.

Ximeng Mao

I am a PhD student in computer science working on machine learning. I am widely interested in applying machine learning and deep learning methods to problems rooted from brain-machine interfaces, and I aim to leverage both deterministic and stochastic models for generic solutions. My current research focuses on meta learning for few-shot adaptation of personalized models and foundation models for neural decoding. My other interests include probabilistic inference, reinforcement learning and medical image analysis.