Leveraging weights signals -- Predicting and improving generalizability in reinforcement learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Moulin, Olivier, Francois-lavet, Vincent, Elbers, Paul, Hoogendoorn, Mark
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918217850552320
author Moulin, Olivier
Francois-lavet, Vincent
Elbers, Paul
Hoogendoorn, Mark
author_facet Moulin, Olivier
Francois-lavet, Vincent
Elbers, Paul
Hoogendoorn, Mark
contents Generalizability of Reinforcement Learning (RL) agents (ability to perform on environments different from the ones they have been trained on) is a key problem as agents have the tendency to overfit to their training environments. In order to address this problem and offer a solution to increase the generalizability of RL agents, we introduce a new methodology to predict the generalizability score of RL agents based on the internal weights of the agent's neural networks. Using this prediction capability, we propose some changes in the Proximal Policy Optimization (PPO) loss function to boost the generalization score of the agents trained with this upgraded version. Experimental results demonstrate that our improved PPO algorithm yields agents with stronger generalizability compared to the original version.
format Preprint
id arxiv_https___arxiv_org_abs_2511_20234
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Leveraging weights signals -- Predicting and improving generalizability in reinforcement learning
Moulin, Olivier
Francois-lavet, Vincent
Elbers, Paul
Hoogendoorn, Mark
Machine Learning
Artificial Intelligence
Generalizability of Reinforcement Learning (RL) agents (ability to perform on environments different from the ones they have been trained on) is a key problem as agents have the tendency to overfit to their training environments. In order to address this problem and offer a solution to increase the generalizability of RL agents, we introduce a new methodology to predict the generalizability score of RL agents based on the internal weights of the agent's neural networks. Using this prediction capability, we propose some changes in the Proximal Policy Optimization (PPO) loss function to boost the generalization score of the agents trained with this upgraded version. Experimental results demonstrate that our improved PPO algorithm yields agents with stronger generalizability compared to the original version.
title Leveraging weights signals -- Predicting and improving generalizability in reinforcement learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2511.20234