Investigating Action Encodings in Recurrent Neural Networks in Reinforcement Learning.
Matthew SchlegelVolodymyr TkachukAdam M. WhiteMartha WhitePublished in: Trans. Mach. Learn. Res. (2023)
Keyphrases
- recurrent neural networks
- reinforcement learning
- action selection
- neural network
- partially observable domains
- action space
- neural model
- complex valued
- reservoir computing
- recurrent networks
- reward shaping
- feed forward
- state space
- feedforward neural networks
- echo state networks
- artificial neural networks
- state action
- cascade correlation
- transition model
- markov decision processes
- function approximation
- learning algorithm
- model free
- agent learns
- optimal control
- expert systems
- knowledge base
- machine learning
- real time
- genetic algorithm
- long term
- optimal policy