Average reward reinforcement learning: foundations, algorithms, and empirical results
Machine Learning - Special issue on reinforcement learning
Learning to Predict by the Methods of Temporal Differences
Machine Learning
Accelerating reinforcement learning through implicit imitation
Journal of Artificial Intelligence Research
Paper: Human dynamics in man-machine systems
Automatica (Journal of IFAC)
Hi-index | 22.14 |
The goal of average reward reinforcement learning is to maximize the long-term average rewards of a generic system. This coincides with the design objective of the control performance standards (CPS) which were established to improve the long-term performance of an automatic generation controller (AGC) used for real-time control of interconnected power systems. In this paper, a novel R(@l) imitation learning (R(@l)IL) method based on the average reward optimality criterion is presented to develop an optimal AGC under the CPS. This R(@l)IL-based AGC can operate online in real-time with high CPS compliances and fast convergence rate in the imitation pre-learning process. Its capability to learn the control behaviors of the existing AGC by observing system variations enable it to overcome the serious defect in the applicability of conventional RL controllers, in which an accurate power system model is required for the offline pre-learning process, and significantly enhance the learning efficiency and control performance for power generation control in various power system operation scenarios.