Abstract
We propose an approach to extract explanations from a trained reinforcement learning agent. Our analysis over three RMSA environment variations shows how the agent uses the input information, increasing our understanding of its learned policy.
© 2024 The Author(s)
PDF ArticleMore Like This
Xiaoliang Chen, Jiannan Guo, Zuqing Zhu, Roberto Proietti, Alberto Castro, and S. J. B. Yoo
W4F.2 Optical Fiber Communication Conference (OFC) 2018
Michael Doherty and Alejandra Beghelli
Th2A.21 Optical Fiber Communication Conference (OFC) 2024
Maria Freire-Hermelo, Antoine Lavignotte, and Catherine Lepers
NeF2B.4 Photonic Networks and Devices (Networks) 2021