VUSFA:Variational Universal Successor Features Approximator

Shamane Siriwardhana, Rivindu Weerasakera, Denys J.C. Matthies, Suranga Nanayakkara

Abstract: In this paper, we show how novel transfer reinforcement learning techniques can be applied to the complex task of target-driven navigation using the photorealisticAI2THOR simulator. Specifically, we build on the concept of Universal SuccessorFeatures with an A3C agent. We introduce the novel architectural1contribution of a Successor Feature Dependent Policy (SFDP) and adopt the concept of VariationalInformation Bottlenecks to achieve state of the art performance.VUSFA, our final architecture, is a straightforward approach that can be implemented using our open source repository. Our approach is generalizable, showed greater stability in training, and outperformed recent approaches in terms of transfer learning ability.

Keywords: Universal Successor Features, Successor Features, Model Free Deep Reinforcement Learning

TL;DR: We present an improved version of Universal Successor Features based DRL method which can improve the transfer learning of agents.