A DPDK-Based Acceleration Method for Experience Sampling of Distributed Reinforcement Learning.
Masaki FurukawaHiroki MatsutaniPublished in: CoRR (2021)
Keyphrases
- prior knowledge
- reinforcement learning
- preprocessing
- high accuracy
- significant improvement
- experimental evaluation
- cost function
- pairwise
- computational complexity
- sampling strategy
- distributed environment
- parameter space
- model free
- segmentation method
- sampling methods
- input data
- data sets
- support vector machine
- dynamic programming
- evolutionary algorithm
- cooperative
- multi agent
- similarity measure
- machine learning