Enregistré dans:
Détails bibliographiques
Auteurs principaux: Voelcker, Claas, Kastner, Tyler, Gilitschenski, Igor, Farahmand, Amir-massoud
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2406.17718
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916300116197376
author Voelcker, Claas
Kastner, Tyler
Gilitschenski, Igor
Farahmand, Amir-massoud
author_facet Voelcker, Claas
Kastner, Tyler
Gilitschenski, Igor
Farahmand, Amir-massoud
contents We investigate the impact of auxiliary learning tasks such as observation reconstruction and latent self-prediction on the representation learning problem in reinforcement learning. We also study how they interact with distractions and observation functions in the MDP. We provide a theoretical analysis of the learning dynamics of observation reconstruction, latent self-prediction, and TD learning in the presence of distractions and observation functions under linear model assumptions. With this formalization, we are able to explain why latent-self prediction is a helpful \emph{auxiliary task}, while observation reconstruction can provide more useful features when used in isolation. Our empirical analysis shows that the insights obtained from our learning dynamics framework predicts the behavior of these loss functions beyond the linear model assumption in non-linear neural networks. This reinforces the usefulness of the linear model framework not only for theoretical analysis, but also practical benefit for applied problems.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17718
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle When does Self-Prediction help? Understanding Auxiliary Tasks in Reinforcement Learning
Voelcker, Claas
Kastner, Tyler
Gilitschenski, Igor
Farahmand, Amir-massoud
Machine Learning
We investigate the impact of auxiliary learning tasks such as observation reconstruction and latent self-prediction on the representation learning problem in reinforcement learning. We also study how they interact with distractions and observation functions in the MDP. We provide a theoretical analysis of the learning dynamics of observation reconstruction, latent self-prediction, and TD learning in the presence of distractions and observation functions under linear model assumptions. With this formalization, we are able to explain why latent-self prediction is a helpful \emph{auxiliary task}, while observation reconstruction can provide more useful features when used in isolation. Our empirical analysis shows that the insights obtained from our learning dynamics framework predicts the behavior of these loss functions beyond the linear model assumption in non-linear neural networks. This reinforces the usefulness of the linear model framework not only for theoretical analysis, but also practical benefit for applied problems.
title When does Self-Prediction help? Understanding Auxiliary Tasks in Reinforcement Learning
topic Machine Learning
url https://arxiv.org/abs/2406.17718