Addressing Function Approximation Error in Actor-Critic Methods

In value-based reinforcement learning methods such as deep Q-learning, function approximation errors are known to lead to overestimated value estimates and suboptimal policies. We show that this problem persists in an actor-critic setting and propose novel mechanisms to minimize its effects on both the actor and the critic... (read more)

PDF Abstract ICML 2018 PDF ICML 2018 Abstract

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods used in the Paper

Double Q-learning
Off-Policy TD Control
Experience Replay
Replay Memory
Dense Connections
Feedforward Networks
Activation Functions
Stochastic Optimization
Target Policy Smoothing
Clipped Double Q-learning
Off-Policy TD Control
Policy Gradient Methods