Back to Search Start Over

Map-based Experience Replay: A Memory-Efficient Solution to Catastrophic Forgetting in Reinforcement Learning

Authors :
Hafez, Muhammad Burhan
Immisch, Tilman
Weber, Tom
Wermter, Stefan
Source :
Frontiers in Neurorobotics 17:1127642 (2023)
Publication Year :
2023

Abstract

Deep Reinforcement Learning agents often suffer from catastrophic forgetting, forgetting previously found solutions in parts of the input space when training on new data. Replay Memories are a common solution to the problem, decorrelating and shuffling old and new training samples. They naively store state transitions as they come in, without regard for redundancy. We introduce a novel cognitive-inspired replay memory approach based on the Grow-When-Required (GWR) self-organizing network, which resembles a map-based mental model of the world. Our approach organizes stored transitions into a concise environment-model-like network of state-nodes and transition-edges, merging similar samples to reduce the memory size and increase pair-wise distance among samples, which increases the relevancy of each sample. Overall, our paper shows that map-based experience replay allows for significant memory reduction with only small performance decreases.

Details

Database :
arXiv
Journal :
Frontiers in Neurorobotics 17:1127642 (2023)
Publication Type :
Report
Accession number :
edsarx.2305.02054
Document Type :
Working Paper
Full Text :
https://doi.org/10.3389/fnbot.2023.1127642