CRC-RL: A Novel Visual Feature Representation Architecture for Unsupervised Reinforcement Learning

TL;DR

CRC-RL introduces a novel visual feature learning architecture using a combined loss function to enhance reinforcement learning performance, outperforming existing methods on complex control tasks.

Contribution

The paper proposes CRC-RL, a new architecture with a combined loss function for improved visual feature learning in unsupervised RL, setting a new benchmark.

Findings

Outperforms state-of-the-art on DeepMind control suite

Provides insights into feature learning via latent visualization

Demonstrates effectiveness of combined loss in RL environments

Abstract

This paper addresses the problem of visual feature representation learning with an aim to improve the performance of end-to-end reinforcement learning (RL) models. Specifically, a novel architecture is proposed that uses a heterogeneous loss function, called CRC loss, to learn improved visual features which can then be used for policy learning in RL. The CRC-loss function is a combination of three individual loss functions, namely, contrastive, reconstruction and consistency loss. The feature representation is learned in parallel to the policy learning while sharing the weight updates through a Siamese Twin encoder model. This encoder model is augmented with a decoder network and a feature projection network to facilitate computation of the above loss components. Through empirical analysis involving latent feature visualization, an attempt is made to provide an insight into the role played by this loss function in learning new action-dependent features and how they are linked to the complexity of the problems being solved. The proposed architecture, called CRC-RL, is shown to outperform the existing state-of-the-art methods on the challenging Deep mind control suite environments by a significant margin thereby creating a new benchmark in this field.