Adapting to Reward Progressivity via Spectral Reinforcement Learning.
Michael DannJohn ThangarajahPublished in: ICLR (2021)
Keyphrases
- reinforcement learning
- function approximation
- state space
- model free
- eligibility traces
- learning agent
- reinforcement learning algorithms
- spectral analysis
- temporal difference
- optimal policy
- machine learning
- markov decision processes
- transfer learning
- hyperspectral
- learning algorithm
- hidden markov models
- total reward
- reinforcement learning methods
- multi agent systems
- learning capabilities
- multi agent
- learning classifier systems
- state action
- reward function
- spectral features
- policy gradient
- multi agent reinforcement learning
- agent learns
- spectral images
- neural network
- long run
- reward shaping