Bootstrap Latent-Predictive Representations for Multitask Reinforcement Learning.
Zhaohan Daniel GuoBernardo Ávila PiresBilal PiotJean-Bastien GrillFlorent AltchéRémi MunosMohammad Gheshlaghi AzarPublished in: CoRR (2020)
Keyphrases
- multi task
- reinforcement learning
- transfer learning
- learning problems
- multitask learning
- multi task learning
- learning tasks
- gaussian processes
- latent variables
- multi class
- multiple tasks
- cross validation
- multiclass support vector machines
- feature selection
- learning algorithm
- markov random field
- data sets
- action recognition
- kernel methods
- semi supervised learning
- supervised learning
- learning process
- bayesian networks
- machine learning
- maximum entropy discrimination
- data mining