Loading...

Reinforcement learning based on active learning method

Sagha, H ; Sharif University of Technology | 2008

414 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/IITA.2008.565
  3. Publisher: 2008
  4. Abstract:
  5. In this paper, a new reinforcement learning approach is proposed which is based on a powerful concept named Active Learning Method (ALM) in modeling. ALM expresses any multi-input-single-output system as a fuzzy combination of some single-input-singleoutput systems. The proposed method is an actor-critic system similar to Generalized Approximate Reasoning based Intelligent Control (GARIC) structure to adapt the ALM by delayed reinforcement signals. Our system uses Temporal Difference (TD) learning to model the behavior of useful actions of a control system. The goodness of an action is modeled on Reward-Penalty-Plane. IDS planes will be updated according to this plane. It is shown that the system can learn with a predefined fuzzy system or without it (through random actions). © 2008 IEEE
  6. Keywords:
  7. Active learning methods ; Actor critics ; Approximate reasonings ; Fuzzy combinations ; Multi-input-single-output systems ; Reinforcement learning approaches ; Reinforcement signals ; Single-output systems ; System use ; Temporal difference learning ; Education ; Feedback control ; Flight control systems ; Information technology ; Reinforcement ; Reinforcement learning ; Fuzzy logic
  8. Source: Proceedings - 2008 2nd International Symposium on Intelligent Information Technology Application, IITA 2008, 21 December 2008 through 22 December 2008, Shanghai ; Volume 2 , 2008 , Pages 598-602 ; 9780769534978 (ISBN)
  9. URL: https://ieeexplore.ieee.org/abstract/document/4739834