Palash Chatterjee
Papers
1
Total Citations
2
H-Index
1
About
Palash Chatterjee is a researcher advancing the frontier of AI planning under uncertainty, with a focus on probabilistic reasoning in continuous domains. His most notable contribution is the development of **DiSProD (Differentiable Symbolic Propagation of Distributions for Planning)**, a novel online planner introduced in his 2023 work. DiSProD addresses a critical challenge: planning in environments with probabilistic transitions across continuous state and action spaces. By constructing a symbolic graph that captures the distribution of future trajectories conditioned on a policy, and leveraging independence assumptions, DiSProD enables efficient, differentiable propagation of uncertainty. This approach bridges the gap between symbolic reasoning and deep learning, allowing for gradient-based optimization in stochastic settings. Though early in its citation trajectory, DiSProD represents a significant methodological leap, offering a principled framework for robust decision-making under uncertainty. Chatterjee’s work is particularly impactful for robotics, autonomous systems, and reinforcement learning, where handling continuous, stochastic environments is paramount. His research stands out for its elegant synthesis of symbolic AI and probabilistic computation, promising scalable solutions for real-world planning problems.
Research Focus
Key Achievements
Top Papers
- 1