Rare event sampling in high-dimensional dynamical systems and analysis of extreme events
| Project/Area Number |
25330299
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Soft computing
|
|---|
| Research Institution | The Institute of Statistical Mathematics |
|---|
Principal Investigator |
Iba Yukito 統計数理研究所, 大学共同利用機関等の部局等, 教授 (30213200)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2015: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2014: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2013: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
|
|---|
| Keywords | レアイベント / 極端事象 / モンテカルロ法 / 確率計算 / シミュレーション / モンテカルロ / 確率 / 時間逆転 / アルゴリズム / レアイベント・サンプリング / インポータンス・サンプリング / 逐次モンテカルロ法 / 確率台風モデル |
|---|
| Outline of Final Research Achievements |
Rare event sampling in nonlinear stochastic dynamical systems is studied. We developed a "time-reversed simulation method" , by which we can efficiently generate rare events and estimate unbiasedly their probabilities. The key idea of the proposed method is to generate paths (trajectories) from the target event (a rare event such as an exact hit of a big typhoon to Tokyo) to initial states; this can improve the efficiency of the simulation when the target region is small. Using sequential importance sampling algorithm (SIS), unbiased estimates of the probabilities of rare events are calculated without much increase of computational burden. The proposed method can be applied to the analysis of risks of extreme events.
|
Report
(4 results)
Research Products
(11 results)
