Scholar
Rishav
Google Scholar ID: DlT4loUAAAAJ
Mila
Citations & Impact
Citations
57
H-index
2
i10-index
2
Publications
7
Co-authors
4
list available
Resume
Academic Achievements
- Paper 'Behavior discovery and attribution for explainable RL' accepted to TMLR 2025.
- Paper 'Handling delays in RL' accepted to ICLR 2025.
- Paper 'KD-LoRA' accepted to NeurIPS ENLSP Workshop.
- Presented 'Behavioral Suite Analysis of Self-Supervised Learning in Atari' at RLVG workshop, RLC 2025.
- Authored technical blogs on distributed JAX training, real-time RL, and CUDA-accelerated DQN.
Background
- An engineer at heart driven by the challenge of building machine learning systems that work reliably at scale.
- Currently researching real-time and explainable reinforcement learning at Mila.
- Long-term goal is to develop trustworthy systems that can learn efficiently from feedback, moving beyond today’s sample-inefficient models.
- Research interests include Offline RL, mechanistic interpretability, reasoning and human-inspired learning, and real-time decision making.
- Overall research vision is to move from brittle, opaque models toward principled algorithms that are mechanistically interpretable, sample-efficient, and reliable.
