Memory-Augmented Reinforcement Learning for Image-Goal Navigation

TL;DR

This paper introduces a memory-augmented reinforcement learning method for image-goal navigation that outperforms previous approaches by effectively using episodic memory and attention mechanisms with only RGB input.

Contribution

The work presents a novel attention-based end-to-end model with episodic memory for navigation, trained with a self-supervised state-embedding network, achieving state-of-the-art results.

Findings

Outperforms previous methods on Gibson dataset

Achieves high navigation accuracy using only RGB input

Establishes new state of the art in image-goal navigation

Abstract

In this work, we present a memory-augmented approach for image-goal navigation. Earlier attempts, including RL-based and SLAM-based approaches have either shown poor generalization performance, or are heavily-reliant on pose/depth sensors. Our method is based on an attention-based end-to-end model that leverages an episodic memory to learn to navigate. First, we train a state-embedding network in a self-supervised fashion, and then use it to embed previously-visited states into the agent's memory. Our navigation policy takes advantage of this information through an attention mechanism. We validate our approach with extensive evaluations, and show that our model establishes a new state of the art on the challenging Gibson dataset. Furthermore, we achieve this impressive performance from RGB input alone, without access to additional information such as position or depth, in stark contrast to related work.