PhD candidate · McGill University & Mila
I’m Priyesh Vijayan. I study reinforcement learning and representation learning, with a focus on how structure supports generalization and decision-making.
About me
I’m a PhD candidate in Computer Science at McGill University and Mila, advised by Doina Precup and Samira E. Kahou. Previously, I completed my MS by Research at IIT Madras under the supervision of Balaraman Ravindran, where I also worked as a researcher at the Robert Bosch Centre for Data Science and AI and the RISE Lab.
My research focuses on reinforcement learning and representation learning, particularly on exploiting the underlying structure of learning problems to improve representations, generalization, and decision-making. My PhD research investigates spectral representations of environment dynamics and their applications to value function approximation and exploration in deep reinforcement learning. My earlier work explored graph neural networks, graph representation learning, and learning from structured data.
More broadly, I am interested in large language models (LLMs), reasoning, and AI for Science. I am particularly interested in how advances in representation learning and reinforcement learning can contribute to more capable, generalizable AI systems and accelerate scientific discovery.
Education, experience & talks →Research contributions
Learning Laplacian representations of environment dynamics when the graph structure is not given, and using them for value function approximation and exploration in deep reinforcement learning.
Explore reinforcement learning →Learning from graph-structured data through multi-hop propagation, local structure, and multiple relations, alongside systematic benchmarking and analysis of graph representations.
Explore graph learning →Using relational structure for dialogue response generation, dynamic scene understanding, vehicle behavior classification, and influence maximization in unknown social networks.
Explore applications →Selected work
Fortieth Annual Conference on Neural Information Processing Systems, NeurIPS'26
Reinforcement Learning Beyond Rewards Workshop, RLC 2026
Inductive Biases in Reinforcement Learning Workshop, RLC 2025
Recent updates
“Addressing Sparse-Rewards in RL with Scalable Hierarchical Novel Eigen Options” accepted at NeurIPS 2026; also presented as a spotlight at the RLC workshop on Reinforcement Learning Beyond Rewards.
Recognized as a Top Reviewer at NeurIPS.
Presented work on Laplacian representations for value function approximation at the RLC workshop on Inductive Biases in Reinforcement Learning.
Get in touch
For research conversations and collaboration, reach me at priyesh@hotmail.co.in.