Towards an Understanding of Default Policies in Multitask Policy Optimization.
Ted MoskovitzMichael ArbelJack Parker-HolderAldo PacchianoPublished in: AISTATS (2022)
Keyphrases
- multi task
- optimal policy
- multitask learning
- multi task learning
- control policies
- learning tasks
- markov decision process
- management policies
- multiple tasks
- transfer learning
- reinforcement learning
- gaussian processes
- allocation policy
- maximum margin
- markov decision processes
- feature space
- reward function
- partially observable markov decision processes
- privacy policies
- multi class
- state space
- active learning
- pairwise