Sequence Compression Speeds Up Credit Assignment in Reinforcement Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ramesh, Aditya A., Young, Kenny, Kirsch, Louis, Schmidhuber, Jürgen
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910470263275520
author Ramesh, Aditya A.
Young, Kenny
Kirsch, Louis
Schmidhuber, Jürgen
author_facet Ramesh, Aditya A.
Young, Kenny
Kirsch, Louis
Schmidhuber, Jürgen
contents Temporal credit assignment in reinforcement learning is challenging due to delayed and stochastic outcomes. Monte Carlo targets can bridge long delays between action and consequence but lead to high-variance targets due to stochasticity. Temporal difference (TD) learning uses bootstrapping to overcome variance but introduces a bias that can only be corrected through many iterations. TD($λ$) provides a mechanism to navigate this bias-variance tradeoff smoothly. Appropriately selecting $λ$ can significantly improve performance. Here, we propose Chunked-TD, which uses predicted probabilities of transitions from a model for computing $λ$-return targets. Unlike other model-based solutions to credit assignment, Chunked-TD is less vulnerable to model inaccuracies. Our approach is motivated by the principle of history compression and 'chunks' trajectories for conventional TD learning. Chunking with learned world models compresses near-deterministic regions of the environment-policy interaction to speed up credit assignment while still bootstrapping when necessary. We propose algorithms that can be implemented online and show that they solve some problems much faster than conventional TD($λ$).
format Preprint
id arxiv_https___arxiv_org_abs_2405_03878
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Sequence Compression Speeds Up Credit Assignment in Reinforcement Learning
Ramesh, Aditya A.
Young, Kenny
Kirsch, Louis
Schmidhuber, Jürgen
Machine Learning
Artificial Intelligence
Temporal credit assignment in reinforcement learning is challenging due to delayed and stochastic outcomes. Monte Carlo targets can bridge long delays between action and consequence but lead to high-variance targets due to stochasticity. Temporal difference (TD) learning uses bootstrapping to overcome variance but introduces a bias that can only be corrected through many iterations. TD($λ$) provides a mechanism to navigate this bias-variance tradeoff smoothly. Appropriately selecting $λ$ can significantly improve performance. Here, we propose Chunked-TD, which uses predicted probabilities of transitions from a model for computing $λ$-return targets. Unlike other model-based solutions to credit assignment, Chunked-TD is less vulnerable to model inaccuracies. Our approach is motivated by the principle of history compression and 'chunks' trajectories for conventional TD learning. Chunking with learned world models compresses near-deterministic regions of the environment-policy interaction to speed up credit assignment while still bootstrapping when necessary. We propose algorithms that can be implemented online and show that they solve some problems much faster than conventional TD($λ$).
title Sequence Compression Speeds Up Credit Assignment in Reinforcement Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2405.03878