Multi-task Deep Reinforcement Learning with Evolutionary Algorithm and Policy Gradients Method in 3D Control Tasks.
Shota ImaiYuichi SeiYasuyuki TaharaRyohei OriharaAkihiko OhsugaPublished in: BCD (2019)
Keyphrases
- multi task
- evolutionary algorithm
- reinforcement learning
- neural network
- similarity measure
- policy search
- control policy
- multiple tasks
- transfer learning
- feature set
- active learning
- mutual information
- unsupervised learning
- theoretical analysis
- cross validation
- supervised learning
- correlation coefficient
- classification accuracy
- multi task learning
- prior knowledge
- data sets