C2RL: Convolutional-Contrastive Learning for Reinforcement Learning Based on Self-Pretraining for Strong Augmentation.
Sanghoon ParkJihun KimHan-You JeongTae-Kyoung KimJinwoo YooPublished in: Sensors (2023)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- autonomous learning
- state space
- learning problems
- learning agents
- rl algorithms
- eligibility traces
- temporal difference learning
- learned knowledge
- reinforcement learning algorithms
- learning tasks
- markov decision processes
- supervised learning
- neural network
- continuous state
- active learning
- reinforcement learning methods
- optimal control
- function approximation
- transfer learning
- adaptive control
- learning mechanism
- optimal policy
- learning agent
- model free
- mobile robot
- dynamic programming
- learning systems
- multi agent
- policy search
- hierarchical reinforcement learning