Portrait de Arushi Jain n'est pas disponible

Arushi Jain

Doctorat - McGill University
Superviseur⋅e principal⋅e


Towards Painless Policy Optimization for Constrained MDPs
Arushi Jain
Sharan Vaswani
Reza Babanezhad Harikandeh
Csaba Szepesvari
We study policy optimization in an infinite horizon, …
Variance Penalized On-Policy and Off-Policy Actor-Critic
Arushi Jain
Gandharv Patil
Ayush Jain
Safe option-critic: learning safety in the option-critic architecture
Abstract Designing hierarchical reinforcement learning algorithms that exhibit safe behaviour is not only vital for practical applications b… (voir plus)ut also facilitates a better understanding of an agent’s decisions. We tackle this problem in the options framework (Sutton, Precup & Singh, 1999), a particular way to specify temporally abstract actions which allow an agent to use sub-policies with start and end conditions. We consider a behaviour as safe that avoids regions of state space with high uncertainty in the outcomes of actions. We propose an optimization objective that learns safe options by encouraging the agent to visit states with higher behavioural consistency. The proposed objective results in a trade-off between maximizing the standard expected return and minimizing the effect of model uncertainty in the return. We propose a policy gradient algorithm to optimize the constrained objective function. We examine the quantitative and qualitative behaviours of the proposed approach in a tabular grid world, continuous-state puddle world, and three games from the Arcade Learning Environment: Ms. Pacman, Amidar, and Q*Bert. Our approach achieves a reduction in the variance of return, boosts performance in environments with intrinsic variability in the reward structure, and compares favourably both with primitive actions and with risk-neutral options.