Loss- and Reward-Weighting for Efficient Distributed Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Holen, Martin, Andersen, Per-Arne, Knausgård, Kristian Muri, Goodwin, Morten
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910568364900352
author Holen, Martin
Andersen, Per-Arne
Knausgård, Kristian Muri
Goodwin, Morten
author_facet Holen, Martin
Andersen, Per-Arne
Knausgård, Kristian Muri
Goodwin, Morten
contents This paper introduces two learning schemes for distributed agents in Reinforcement Learning (RL) environments, namely Reward-Weighted (R-Weighted) and Loss-Weighted (L-Weighted) gradient merger. The R/L weighted methods replace standard practices for training multiple agents, such as summing or averaging the gradients. The core of our methods is to scale the gradient of each actor based on how high the reward (for R-Weighted) or the loss (for L-Weighted) is compared to the other actors. During training, each agent operates in differently initialized versions of the same environment, which gives different gradients from different actors. In essence, the R-Weights and L-Weights of each agent inform the other agents of its potential, which again reports which environment should be prioritized for learning. This approach of distributed learning is possible because environments that yield higher rewards, or low losses, have more critical information than environments that yield lower rewards or higher losses. We empirically demonstrate that the R-Weighted methods work superior to the state-of-the-art in multiple RL environments.
format Preprint
id arxiv_https___arxiv_org_abs_2304_12778
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Loss- and Reward-Weighting for Efficient Distributed Reinforcement Learning
Holen, Martin
Andersen, Per-Arne
Knausgård, Kristian Muri
Goodwin, Morten
Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
This paper introduces two learning schemes for distributed agents in Reinforcement Learning (RL) environments, namely Reward-Weighted (R-Weighted) and Loss-Weighted (L-Weighted) gradient merger. The R/L weighted methods replace standard practices for training multiple agents, such as summing or averaging the gradients. The core of our methods is to scale the gradient of each actor based on how high the reward (for R-Weighted) or the loss (for L-Weighted) is compared to the other actors. During training, each agent operates in differently initialized versions of the same environment, which gives different gradients from different actors. In essence, the R-Weights and L-Weights of each agent inform the other agents of its potential, which again reports which environment should be prioritized for learning. This approach of distributed learning is possible because environments that yield higher rewards, or low losses, have more critical information than environments that yield lower rewards or higher losses. We empirically demonstrate that the R-Weighted methods work superior to the state-of-the-art in multiple RL environments.
title Loss- and Reward-Weighting for Efficient Distributed Reinforcement Learning
topic Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2304.12778