Grant-in-Aid for General Scientific Research (B)
|Allocation Type||Single-year Grants|
|Research Institution||TOKYO INSTITUTE OF TECHNOLOGY|
KOBAYASHI Shigenobu Tokyo Institute of Technology, Inter disciprinary Graduate School of Science and Engineering, Professor, 大学院・総合理工学研究所, 教授 (40016697)
YAMAMURA Masayuki Tokyo Institute of Technology, Inter disciprinary Graduate School of Science and, 大学院・総合理工学研究所, 助手 (00220442)
|Project Period (FY)
1993 – 1994
Completed(Fiscal Year 1994)
|Budget Amount *help
¥5,600,000 (Direct Cost : ¥5,600,000)
Fiscal Year 1994 : ¥2,100,000 (Direct Cost : ¥2,100,000)
Fiscal Year 1993 : ¥3,500,000 (Direct Cost : ¥3,500,000)
|Keywords||genetic algorithm / role of coding and crossover / combinatorial optimization / multiobjective optimization / reinforcement learning / profit sharing / k-certain exploration / perceptual aliasing / サブツアー交換交叉 / 経験強化型学習 / 環境同定型学習 / マルコフ解析 / 数理生態学的解析 / だまし境界定理 / Q-learing / 割引き勾配法|
Research results through two years are summarized as follows ;
(1) Research results in evolutionary computation
1) An analysis on the role of crossover operators in genetic algorithms
We made a mathematical analysis on the effect of crossover operators to search better solutions with genetic algorithms, and showed the "deceptive boundary theorem."
2) A proposal of evaluation criteria on a coding-crossover
We proposed four criteria of completeness, soundness, non-redundancy and character preservingness to evaluate a model formulation.
3) A proposal and applications of the subtour exchange crossover as a character preserving operation
We proposed a character preserving operation named the subtour exchange crossover in the domain of ordering problems.
4) A proposal of genetic algorithms for multiobjective optimization problems
We established a methodology for genetic algorithms to generate the pareto optimal set, which is the rational solution for a multiobjective optimization.
(2) Research results
in adaptive computation
1) A proposal of a framework and a categorization for reinforcement learning
We defined a transparent framework for reinforcement learning, and classified existing researches according to the class of the environment and the orientation of the approach.
2) An analysis on the rationality of the exploitation oriented reinforcement learning
We made a mathematical analysis on the rationality of the sharing functions in the profit sharing method, and showed two kinds of rationality theorem.
3) A proposal and an extention of the k-certain exploration method
We proposed a exploration oriented action selection strategy named the k-certain exploration method to efficiently identify an unknown environment.
4) A proposal of a reinforcement learning method based on a hillclimbing of the expected rewards
We proposed an incremental reinforcement learning method, which performs a hillclimbing along the gradient of the expected rewards.
5) A proposal of a learning method under perceptual aliasing
We proposed a prediction model to foresee the state transition under incomplete deceptive perception, and developed a learning method to construct a prediction model with trial and error. Less