Adapting to Reward Progressivity via Spectral Reinforcement Learning.
Michael DannJohn ThangarajahPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- eligibility traces
- state space
- reward function
- model free
- average reward
- temporal difference
- spectral images
- markov decision processes
- learning algorithm
- learning capabilities
- dynamic programming
- multi agent
- optimal policy
- learning problems
- transfer learning
- spectral analysis
- policy iteration
- reinforcement learning methods
- hidden markov models
- partially observable environments
- reward shaping
- total reward
- machine learning
- policy search
- multi agent reinforcement learning
- spectral data
- temporal difference learning
- optimal control
- control policy
- learning agent
- supervised learning
- learning classifier systems