Less is more - the Dispatcher/ Executor principle for multi-task Reinforcement Learning.
Martin A. RiedmillerTim HertweckRoland HafnerPublished in: CoRR (2023)
Keyphrases
- multi task
- reinforcement learning
- transfer learning
- learning problems
- multi task learning
- learning tasks
- multitask learning
- multiple tasks
- multi class
- metric learning
- feature selection
- sparse learning
- gaussian processes
- supervised learning
- learning algorithm
- machine learning
- information gain
- training data
- kernel methods
- learning experience
- multiple kernel learning
- similarity measure