ReLeQ: A Reinforcement Learning Approach for Deep Quantization of Neural Networks.
Ahmed T. ElthakebPrannoy PilligundlaAmir YazdanbakhshSean KinzerHadi EsmaeilzadehPublished in: CoRR (2018)
Keyphrases
- neural network
- reinforcement learning
- function approximators
- function approximation
- pattern recognition
- artificial neural networks
- learning classifier systems
- state space
- fuzzy logic
- recurrent neural networks
- neural network model
- learning capabilities
- learning algorithm
- fault diagnosis
- multilayer perceptron
- quantization error
- fuzzy systems
- fuzzy neural network
- feed forward
- network architecture
- markov decision processes
- policy search
- genetic algorithm
- neural nets
- multi agent reinforcement learning
- robotic control
- reinforcement learning methods
- quantization step
- deep learning
- competitive learning
- model free
- mobile robot
- expert systems
- machine learning