The Archimedean trap: Why traditional reinforcement learning will probably not yield AGI.
Samuel Allen AlexanderPublished in: J. Artif. Gen. Intell. (2020)
Keyphrases
- reinforcement learning
- action selection
- artificial general intelligence
- intelligent systems
- optimal policy
- reinforcement learning algorithms
- markov decision processes
- state space
- function approximation
- learning environment
- temporal difference
- multi agent
- robotic control
- artificial intelligence
- model free
- dynamic programming
- cognitive model
- action space
- mobile robot
- markov decision process
- lagrangian method