| Project/Area Number |
15K00347
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Soft computing
|
|---|
| Research Institution | Kyoto College of Graduate Studies for Informatics (2017-2019)
Hachinohe Institute of Technology (2015-2016) |
|---|
Principal Investigator |
Takahashi Ryouei 京都情報大学院大学, その他の研究科, 教授 (10347841)
|
|---|
| Project Period (FY) |
2015-10-21 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2018: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2017: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2015: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | ACO / GA / Neural Network / CMA-ES / Classification / Function Optimization / Diversity measurements / 段階的探索空間局所化機能 / Neural Network / Cross Entropy / Stepwise localization / FOP / Real Coded GA / Ant colony optimization / Explore search space / Statistics / Real coded GA / Diversity Measurement / ACO Continuous Domains / iversity Measrement / Changing Crossover / BLX-α / SPX / TDGA / Immune-GA / Diversity Measrement |
|---|
| Outline of Final Research Achievements |
We can recognize that almost of social problems are represented by function minimization problems(FMP). This function is expressed as the square of errors between the predicted values and the observed values, or the distance between the entropy of the real probability distribution and that of the predicted model's probability distribution. FMP is the problem to determine unknown parameters of predicted models to such an extent that the function has its minimum value. In this study, we mainly investigated three optimization methods to solve FMP. They are Genetic algorithms which are based on Darwin's evolutionary theory, Ant Colony optimization which simulates ants' behavior of searching for feeds between nest and foods, and CMA-ES which searches for solutions based on mutation operations. Results of our C experiments say that they are good optimizers to solve FMP. We are successively reporting our experimental results on the international conference such as WCCI, ANTS, and ICMLA.
|
| Academic Significance and Societal Importance of the Research Achievements |
情報通信技術の進展により、世界中に張り巡らされた膨大なコンピュータネットワークシステムを介して、多種多様な人間の要求や行動が各種センサ等により感知・観測できるようになり、世界中どこにいても即座にタイムリィにそれらの情報を収集・蓄積できるようになった。このネットワークに収集蓄積されたビッグデータを解析し、人工知能と人間が共存して社会の問題や産業の問題を解決していく社会Society5.0が望まれている。例えば、心電図、血液検査、レントゲン撮影等の情報から患者の異常を検出する医療診断に人工知能を導入する等である。その際人工知能は、教師あり学習という方法で関数最小値問題を解き、医師の知能を学習する。
|
