More capable RL agents exploit reward misspecifications more often, with phase transitions in behavior, and anomaly detectors can identify misaligned policies.
Reinforcement learn ing in healthcare: A survey
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2022 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
More capable RL agents exploit reward misspecifications more often, with phase transitions in behavior, and anomaly detectors can identify misaligned policies.