ACTER: Diverse and Actionable Counterfactual Sequences for Explaining and Diagnosing RL Policies.
Jasmina GajcinIvana DusparicPublished in: CoRR (2024)
Keyphrases
- optimal policy
- reinforcement learning
- hidden markov models
- control policies
- wide variety
- control policy
- markov decision process
- real world
- state space
- markov decision processes
- learning algorithm
- partially observable markov decision processes
- model free
- learning classifier systems
- model based diagnosis
- sequence alignment
- causal reasoning
- learning agents
- dynamic programming
- semi markov decision process