Reward is not enough: can we liberate AI from the reinforcement learning paradigm?
Vacslav GlukhovPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- artificial intelligence
- state space
- machine learning
- reinforcement learning algorithms
- function approximation
- temporal difference
- eligibility traces
- expert systems
- markov decision processes
- reward function
- dynamic programming
- partially observable environments
- programming paradigms
- optimal policy
- computational intelligence
- knowledge based systems
- learning algorithm
- ai systems
- multi agent
- case based reasoning
- supervised learning
- model free
- average reward
- state action
- ai technologies
- learning process
- lecture notes in artificial intelligence
- learning problems
- robotic control
- policy iteration
- neural network
- intelligent systems
- total reward
- transfer learning
- decision problems
- robot control
- partially observable