Exploring the relationship of reward and punishment in reinforcement learning.
Robert LoweTom ZiemkePublished in: ADPRL (2013)
Keyphrases
- reinforcement learning
- function approximation
- agent receives
- state space
- eligibility traces
- reward function
- reinforcement learning algorithms
- multi agent
- partially observable environments
- model free
- optimal control
- learning capabilities
- optimal policy
- learning process
- markov decision processes
- stochastic approximation
- learning algorithm
- temporal difference
- database
- supervised learning
- learning classifier systems
- dynamic programming
- case study
- information systems
- learning problems
- learning agent
- temporal difference learning
- reinforcement learning methods
- policy gradient
- autonomous learning
- multi agent reinforcement learning
- genetic algorithm
- data sets
- real time