Recently, a reinforcement learning improves its potential by integrating with deep learning, e.g., deep Q-networks (DQN) and AlphaGO proposed by Google DeepMind. To explore new potential of reinforcement learning, this special issue focuses on both reinforcement learning and its hybrid methods (such as machine learning, neural network, evolutionary computation). Research on meta-learning, multitask learning, curriculum learning, intrinsic motivation, reward shaping for a reinforcement learning is also welcome in addition to theoretical and application research.

*Reinforcement learning (including multiagent reinforcement learning and inverse reinforcement learning)
*Hybrid reinforcement learning (e.g., deep reinforcement learning)
*Hybrid machine learning (e.g., evolutionary machine learning)
*Machine learning technology for reinforcement learning (e.g., neural network / evolutionary computation / decision tree / random forest / support vector machine / k-nearest neighbor / self-organizing maps / Bayesian network for reinforcement learning)
*Data mining, classification, regression, pattern extraction/recognition by reinforcement learning and its hybrid methods
*Theory of reinforcement learning and its hybrid methods
*Application of reinforcement learning and its hybrid methods

We welcome researchers to submit their achievement to this special issue as an opportunity for publishing in a high standard international journal.

Regular paper submissions are always welcome!