Navdeep Kumar

Latest

• Non-rectangular Robust MDPs with Normed Uncertainty Sets
• On the Convergence of Single-Timescale Actor-Critic
• Global Convergence of Policy Gradient in Average Reward MDPs