GAN-based Intrinsic Exploration for Sample Efficient Reinforcement Learning

Araştırma sonucu: ???type-name???Konferans makalesibilirkişi

Özet

In this study, we address the problem of efficient exploration in reinforcement learning. Most common exploration approaches depend on random action selection, however these approaches do not work well in environments with sparse or no rewards. We propose Generative Adversarial Network-based Intrinsic Reward Module that learns the distribution of the observed states and sends an intrinsic reward that is computed as high for states that are out of distribution, in order to lead agent to unexplored states. We evaluate our approach in Super Mario Bros for a no reward setting and in Montezuma’s Revenge for a sparse reward setting and show that our approach is indeed capable of exploring efficiently. We discuss a few weaknesses and conclude by discussing future works.

Orijinal dilİngilizce
Sayfa (başlangıç-bitiş)264-272
Sayfa sayısı9
DergiInternational Conference on Agents and Artificial Intelligence
Hacim2
DOI'lar
Yayın durumuYayınlandı - 2022
Etkinlik14th International Conference on Agents and Artificial Intelligence , ICAART 2022 - Virtual, Online
Süre: 3 Şub 20225 Şub 2022

Bibliyografik not

Publisher Copyright:
© 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved.

Finansman

This work is supported by Istanbul Technical University BAP Grant NO: MOA-2019-42321.

FinansörlerFinansör numarası
Istanbul Teknik ÜniversitesiMOA-2019-42321

    Parmak izi

    GAN-based Intrinsic Exploration for Sample Efficient Reinforcement Learning' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

    Alıntı Yap