Reactive Exploration to Cope With Non-Stationarity in Lifelong Reinforcement Learning.
Christian Alexander SteinparzThomas SchmiedFabian PaischerMarius-Constantin DinuVihang Prakash PatilAngela Bitto-NemlingHamid Eghbal-zadehSepp HochreiterPublished in: CoLLAs (2022)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- action selection
- exploration exploitation
- function approximation
- autonomous learning
- non stationary
- model based reinforcement learning
- model free
- reactive agents
- learning scenarios
- multi agent
- active learning
- reinforcement learning algorithms
- temporal difference
- exploration exploitation tradeoff
- state space
- agent architecture
- learning problems
- markov decision problems
- policy search
- transfer learning
- optimal policy
- supervised learning
- robotic control
- learning process
- learning classifier systems
- markov decision processes
- learning algorithm