Understanding Bonus-Based Exploration in Reinforcement Learning
Name
Chen-ericrc-meng-eecs-2021-thesis.pdf
Description
Thesis PDF
Size
10.25 MB
Format
Adobe PDF
Checksum (MD5)
2a46c7cf3bbbb8676ff359ac9ac73c84
Author(s)
Chen, Eric
Advisor(s)
Agrawal, Pulkit
Date Issued
September 2021
Publisher
Massachusetts Institute of Technology
Abstract
Intrinsic reward-based exploration methods have successfully solved challenging sparse reward tasks such as Montezuma’s Revenge. However, these methods have not been widely adopted in reinforcement learning due to inconsistent performance gains across tasks. To better understand the underlying cause of this variability, we evaluate the performance of three major families of exploration methods on a suite of custom environments and video games: prediction error, state visitation and model uncertainty. Our custom environments allow us to study the effect of different environmental features in isolation. Our results reveal that exploration methods can be biased by spurious features such as color, and prioritize different dynamics in specific environments. In particular, we find that prediction-based methods are superior at solving tasks involving controllable dynamics. Furthermore, we find that partial observability can hinder exploration by setting up "curiosity traps" that agents can fall into. Finally, we investigate how various implementation details such as reward design and generation affect an agent’s overall performance.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
In Copyright - Educational Use Permitted
Copyright MIT
Persistent DSpace Link
