Abstract
Reinforcement learning is a type of machine learning in which an agent learns by taking actions in an environment and receiving feedback in the form of reward signals. The objective of the agent is to maximize the cumulative reward over time. However, there are certain challenges that can hinder the learning process. These challenges include difficult-to-explore environments, sparse or uncertain rewards.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A nice visualization of the Montezuma’s Revenge game from OpenAI: https://openai.com/research/reinforcement-learning-with-prediction-based-rewards
References
Deepak Pathak, Pulkit Agrawal, Alexei A. Efros, and Trevor Darrell. Curiosity-driven exploration by self-supervised prediction, 2017.
Yuri Burda, Harrison Edwards, Amos Storkey, and Oleg Klimov. Exploration by random network distillation, 2018.
Adrià Puigdomènech Badia, Bilal Piot, Steven Kapturowski, Pablo Sprechmann, Alex Vitvitskyi, Daniel Guo, and Charles Blundell. Agent57: Outperforming the Atari human benchmark, 2020.
Zhaohan Daniel Guo and Emma Brunskill. Directed exploration for reinforcement learning, 2019.
Enrico Marchesini, Davide Corsi, and Alessandro Farinelli. Benchmarking safe deep reinforcement learning in aquatic navigation, 2021.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature
About this chapter
Cite this chapter
Hu, M. (2023). Curiosity-Driven Exploration. In: The Art of Reinforcement Learning. Apress, Berkeley, CA. https://doi.org/10.1007/978-1-4842-9606-6_13
Download citation
DOI: https://doi.org/10.1007/978-1-4842-9606-6_13
Published:
Publisher Name: Apress, Berkeley, CA
Print ISBN: 978-1-4842-9605-9
Online ISBN: 978-1-4842-9606-6
eBook Packages: Professional and Applied ComputingApress Access BooksProfessional and Applied Computing (R0)