No Free Lunch: Overcoming Reward Gaming in AI Safety Gridworlds
Attributed to:
UKRI Trustworthy Autonomous Systems Node in Verifiability
funded by
SPF
Abstract
No abstract provided
Bibliographic Information
Digital Object Identifier: http://dx.doi.org/10.1007/978-3-030-83906-2_18
Publication URI: http://dx.doi.org/10.1007/978-3-030-83906-2_18
Type: Book Chapter
Book Title: Computer Safety, Reliability, and Security. SAFECOMP 2021 Workshops - DECSoS, MAPSOD, DepDevOps, USDAI, and WAISE, York, UK, September 7, 2021, Proceedings (2021)
Page Reference: 226-238