Decoupled Reinforcement Learning to Stabilise Intrinsically-Motivated Exploration.
Lukas SchäferFilippos ChristianosJosiah P. HannaStefano V. AlbrechtPublished in: AAMAS (2022)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- action selection
- model based reinforcement learning
- function approximation
- exploration exploitation
- autonomous learning
- state space
- exploration exploitation tradeoff
- reinforcement learning algorithms
- data sets
- unknown environments
- neural network
- learning environment
- temporal difference learning
- temporal difference
- learning process
- markov decision processes
- artificial neural networks
- partially observable
- stochastic approximation
- supervised learning
- model free
- interactive exploration
- transition model
- case study
- machine learning
- real world
- robotic control
- active learning