Dynamic Reward-Based Dueling Deep Dyna-Q: Robust Policy Learning in Noisy Environments.
Yangyang ZhaoZhenyu WangKai YinRui ZhangZhenhua HuangPei WangPublished in: AAAI (2020)
Keyphrases
- noisy environments
- reinforcement learning
- learning algorithm
- inverse reinforcement learning
- image processing
- partially observable environments
- prior knowledge
- speech recognition
- learning problems
- noise reduction
- function approximation
- pattern recognition
- multiresolution
- wavelet coefficients
- action selection
- speaker verification