On the consistency of hyper-parameter selection in value-based deep reinforcement learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Obando-Ceron, Johan, Araújo, João G. M., Courville, Aaron, Castro, Pablo Samuel
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913590881026048
author Obando-Ceron, Johan
Araújo, João G. M.
Courville, Aaron
Castro, Pablo Samuel
author_facet Obando-Ceron, Johan
Araújo, João G. M.
Courville, Aaron
Castro, Pablo Samuel
contents Deep reinforcement learning (deep RL) has achieved tremendous success on various domains through a combination of algorithmic design and careful selection of hyper-parameters. Algorithmic improvements are often the result of iterative enhancements built upon prior approaches, while hyper-parameter choices are typically inherited from previous methods or fine-tuned specifically for the proposed technique. Despite their crucial impact on performance, hyper-parameter choices are frequently overshadowed by algorithmic advancements. This paper conducts an extensive empirical study focusing on the reliability of hyper-parameter selection for value-based deep reinforcement learning agents, including the introduction of a new score to quantify the consistency and reliability of various hyper-parameters. Our findings not only help establish which hyper-parameters are most critical to tune, but also help clarify which tunings remain consistent across different training regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17523
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle On the consistency of hyper-parameter selection in value-based deep reinforcement learning
Obando-Ceron, Johan
Araújo, João G. M.
Courville, Aaron
Castro, Pablo Samuel
Machine Learning
Artificial Intelligence
Deep reinforcement learning (deep RL) has achieved tremendous success on various domains through a combination of algorithmic design and careful selection of hyper-parameters. Algorithmic improvements are often the result of iterative enhancements built upon prior approaches, while hyper-parameter choices are typically inherited from previous methods or fine-tuned specifically for the proposed technique. Despite their crucial impact on performance, hyper-parameter choices are frequently overshadowed by algorithmic advancements. This paper conducts an extensive empirical study focusing on the reliability of hyper-parameter selection for value-based deep reinforcement learning agents, including the introduction of a new score to quantify the consistency and reliability of various hyper-parameters. Our findings not only help establish which hyper-parameters are most critical to tune, but also help clarify which tunings remain consistent across different training regimes.
title On the consistency of hyper-parameter selection in value-based deep reinforcement learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2406.17523