Biologically-Based Neural Representations Enable Fast Online Shallow Reinforcement Learning.
Madeleine BartlettTerrence C. StewartJeff OrchardPublished in: CogSci (2022)
Keyphrases
- reinforcement learning
- fitted q iteration
- online learning
- neural network
- distributed representations
- state space
- natural language processing
- function approximation
- multi agent
- information extraction
- machine learning
- network architecture
- model free
- learning process
- active learning
- dynamic programming
- real time
- knowledge base
- function approximators
- internal representations