Distributional Reinforcement Learning with Unconstrained Monotonic Neural Networks.
Thibaut ThéateAntoine WehenkelAdrien BollandGilles LouppeDamien ErnstPublished in: CoRR (2021)
Keyphrases
- neural network
- reinforcement learning
- function approximators
- function approximation
- pattern recognition
- state space
- co occurrence
- artificial neural networks
- back propagation
- machine learning
- reinforcement learning algorithms
- learning process
- fuzzy logic
- neural nets
- robotic control
- neural network model
- genetic algorithm
- learning capabilities
- recurrent neural networks
- learning classifier systems
- self organizing maps
- temporal difference learning
- learning algorithm
- temporal difference
- model free
- feed forward
- supervised learning
- least squares
- multi layer
- fuzzy systems
- associative memory
- activation function
- learning rules
- recognition rate
- multi agent
- radial basis function
- markov decision processes
- hopfield neural network
- dynamical systems
- fault diagnosis