Timezone: »

Convergent Policy Optimization for Safe Reinforcement Learning
Ming Yu · Zhuoran Yang · Mladen Kolar · Zhaoran Wang

Tue Dec 10 05:30 PM -- 07:30 PM (PST) @ East Exhibition Hall B + C #201

We study the safe reinforcement learning problem with nonlinear function approximation, where policy optimization is formulated as a constrained optimization problem with both the objective and the constraint being nonconvex functions. For such a problem, we construct a sequence of surrogate convex constrained optimization problems by replacing the nonconvex functions locally with convex quadratic functions obtained from policy gradient estimators. We prove that the solutions to these surrogate problems converge to a stationary point of the original nonconvex problem. Furthermore, to extend our theoretical results, we apply our algorithm to examples of optimal control and multi-agent reinforcement learning with safety constraints.

Author Information

Ming Yu (The University of Chicago, Booth School of Business)
Zhuoran Yang (Princeton University)
Mladen Kolar (University of Chicago)
Zhaoran Wang (Northwestern University)

More from the Same Authors