Seraph: A Performance-Cost Aware Tuner for Training Reinforcement Learning Model on Serverless Computing.
Jinbo HanXingda WeiRong ChenHaibo ChenPublished in: APSys (2024)
Keyphrases
- reinforcement learning
- formal model
- high level
- similarity measure
- mathematical model
- computational model
- em algorithm
- probabilistic model
- dynamic programming
- total cost
- statistical model
- theoretical analysis
- probability distribution
- genetic algorithm
- active learning
- learning process
- training set
- multi agent
- decision trees
- knowledge base