Adaptive perturbation adversarial training: based on reinforcement learning.
Zhishen NieYing LinSp RenLan ZhangPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- multi agent
- supervised learning
- neural network
- markov decision process
- dynamic programming
- training examples
- feedforward neural networks
- function approximation
- training algorithm
- training phase
- multi agent reinforcement learning
- data sets
- reinforcement learning algorithms
- multi layer perceptron
- markov decision processes
- test set
- learning process
- case study