MQGrad: Reinforcement Learning of Gradient Quantization in Parameter Server.
Guoxin CuiJun XuWei ZengYanyan LanJiafeng GuoXueqi ChengPublished in: ICTIR (2018)
Keyphrases
- reinforcement learning
- policy gradient
- function approximation
- client server
- web server
- model free
- markov decision processes
- machine learning
- multi agent
- state space
- reinforcement learning algorithms
- optimal policy
- database
- learning algorithm
- learning process
- parameter values
- neural network
- temporal difference
- input parameters
- parameter settings
- parameter space
- edge detection
- computational complexity
- transfer learning
- supervised learning
- image gradient
- quantization error
- client server architecture
- special case