Maximum entropy exploration in contextual bandits with neural networks and energy based models.
Adam ElwoodMarco LeonardiAshraf MohamedAlessandro RozzaPublished in: CoRR (2022)
Keyphrases
- maximum entropy
- markov models
- maximum entropy principle
- random fields
- neural network
- joint distribution
- potential functions
- probabilistic logic
- class conditional
- maximum entropy model
- probabilistic model
- transformation based learning
- computer vision
- machine learning
- bregman divergences
- iterative scaling
- principle of maximum entropy
- multi task
- bayesian framework
- statistical models
- conditional random fields
- support vector machine
- prior knowledge
- pairwise