Count-Based Temperature Scheduling for Maximum Entropy Reinforcement Learning.
Dailin HuPieter AbbeelRoy FoxPublished in: CoRR (2021)
Keyphrases
- maximum entropy
- reinforcement learning
- maximum entropy principle
- markov models
- scheduling problem
- principle of maximum entropy
- class conditional
- iterative scaling
- random fields
- probabilistic logic
- maximum entropy model
- conditional random fields
- minimum cross entropy
- state space
- learning algorithm
- learning problems
- markov decision processes
- supervised learning
- training data
- optimal policy
- least squares
- transformation based learning
- feature selection