A3CMal: Generating adversarial samples to force targeted misclassification by reinforcement learning.
Zhiyang FangJunfeng WangJiaxuan GengYingjie ZhouXuan KanPublished in: Appl. Soft Comput. (2021)
Keyphrases
- reinforcement learning
- multi agent
- error rate
- cost sensitive
- data sets
- function approximation
- reinforcement learning algorithms
- state space
- learning algorithm
- dynamic programming
- optimal policy
- machine learning
- misclassification rate
- policy search
- sample set
- sample points
- markov decision processes
- transfer learning
- supervised learning
- misclassification costs
- closed loop
- temporal difference learning
- autonomous learning
- multi agent reinforcement learning