Addressing Sample Complexity in Visual Tasks Using HER and Hallucinatory GANs

TL;DR

This paper enhances reinforcement learning in visual environments by combining hallucinated visual trajectories generated by a GAN with HER, significantly improving sample efficiency in 3D navigation and robotics tasks.

Contribution

It introduces a novel method that uses hallucinated visual trajectories with HER to address sample complexity in visual RL tasks.

Findings

Marked improvement over baselines in 3D navigation tasks

Effective use of GAN-generated visual trajectories

Enhanced sample efficiency in visual reinforcement learning

Abstract

Reinforcement Learning (RL) algorithms typically require millions of environment interactions to learn successful policies in sparse reward settings. Hindsight Experience Replay (HER) was introduced as a technique to increase sample efficiency by reimagining unsuccessful trajectories as successful ones by altering the originally intended goals. However, it cannot be directly applied to visual environments where goal states are often characterized by the presence of distinct visual features. In this work, we show how visual trajectories can be hallucinated to appear successful by altering agent observations using a generative model trained on relatively few snapshots of the goal. We then use this model in combination with HER to train RL agents in visual settings. We validate our approach on 3D navigation tasks and a simulated robotics application and show marked improvement over baselines derived from previous work.