Structure Learning-Based Task Decomposition for Reinforcement Learning in Non-stationary Environments.
Honguk WooGwangpyo YooMinjong YooPublished in: AAAI (2022)
Keyphrases
- structure learning
- reinforcement learning
- transfer learning
- graphical models
- bayesian networks
- conditional independence
- parameter learning
- bayesian network structure learning
- parameter estimation
- markov logic networks
- bayesian network classifiers
- learning algorithm
- machine learning
- dynamic programming
- sample size
- structure learning algorithm
- network structure
- random variables
- class imbalance
- markov networks
- state space
- probabilistic reasoning
- data sets
- unlabeled data
- learning process
- optical flow
- natural language processing
- belief propagation
- maximum likelihood
- labeled data