Goodhart's Law in Reinforcement Learning.
Jacek KarwowskiOliver HaymanXingjian BaiKlaus KiendlhoferCharlie GriffinJoar Max Viktor SkalsePublished in: ICLR (2024)
Keyphrases
- supervised learning
- reinforcement learning
- learning algorithm
- machine learning
- temporal difference
- state space
- function approximation
- model free
- active learning
- reinforcement learning algorithms
- transfer learning
- control problems
- temporal difference learning
- learning process
- continuous state
- multi agent
- learning agents
- policy search
- function approximators
- multi agent reinforcement learning
- legal reasoning
- partially observable
- markov decision processes
- optimal policy
- search algorithm
- case study
- artificial intelligence