Generating Adversarial Examples for Low-Resource NMT via Multi-Reward Reinforcement Learning.
Shuo SunHongxu HouZongheng YangYisong WangNier WuPublished in: ICTAI (2022)
Keyphrases
- reinforcement learning
- multi agent
- function approximation
- model free
- resource allocation
- agent receives
- learning capabilities
- reward function
- eligibility traces
- markov decision processes
- partially observable environments
- reinforcement learning algorithms
- optimal policy
- robotic control
- state space
- multi agent reinforcement learning
- action selection
- genetic algorithm
- total reward
- reinforcement learning methods
- key concepts
- partially observable
- learning classifier systems
- data sets
- learning problems
- transfer learning
- least squares
- learning process
- machine learning