Boosting the Convergence of Reinforcement Learning-based Auto-pruning Using Historical Data.
Jiandong MuMengdi WangFeiwen ZhuJun YangWei LinWei ZhangPublished in: CoRR (2021)
Keyphrases
- historical data
- reinforcement learning
- stochastic approximation
- learning algorithm
- search space
- data mining techniques
- predictive model
- function approximation
- convergence rate
- reinforcement learning algorithms
- demand forecasting
- optimal policy
- pruning method
- stream data
- model free
- feature selection
- machine learning
- power plant
- decision trees
- temporal difference
- pruning algorithm
- markov decision processes
- state space
- support vector machine
- stock price
- dynamic programming