Reinforcement Learning and the Reward Engineering Principle.
Daniel DeweyPublished in: AAAI Spring Symposia (2014)
Keyphrases
- reinforcement learning
- function approximation
- state space
- eligibility traces
- reinforcement learning algorithms
- engineering design
- computer science
- total reward
- optimal policy
- artificial intelligence
- reward function
- model free
- markov decision processes
- software engineering
- temporal difference
- learning algorithm
- temporal difference learning
- engineering problems
- supervised learning
- learning classifier systems
- long run
- mechanical engineering
- partially observable
- partially observable environments
- multi agent reinforcement learning
- policy gradient
- function approximators
- action space
- learning agent
- learning capabilities
- action selection
- transfer learning
- learning process
- multi agent