Poster
Finite-Time Performance Bounds and Adaptive Learning Rate Selection for Two Time-Scale Reinforcement Learning
Harsh Gupta · R. Srikant · Lei Ying
East Exhibition Hall B, C #207
Keywords: [ Reinforcement Learning ] [ Reinforcement Learning and Planning ] [ Markov Decision Processes ] [ Reinforcement Learning and Planning -> Decision and Control; Reinforcement Learning and Planning ]
We study two time-scale linear stochastic approximation algorithms, which can be used to model well-known reinforcement learning algorithms such as GTD, GTD2, and TDC. We present finite-time performance bounds for the case where the learning rate is fixed. The key idea in obtaining these bounds is to use a Lyapunov function motivated by singular perturbation theory for linear differential equations. We use the bound to design an adaptive learning rate scheme which significantly improves the convergence rate over the known optimal polynomial decay rule in our experiments, and can be used to potentially improve the performance of any other schedule where the learning rate is changed at pre-determined time instants.
Live content is unavailable. Log in and register to view live content