본문영역
|
융대원 지능형융합시스템학과 세미나 안내 - 연사 : 이선환 박사 (Lattice Engines Inc.)
작성자 : 관리자작성일 :2012-02-20 17:41:16조회수 : 3562
: 전체
|
|||||
|---|---|---|---|---|---|
|
Title : Parallel Stochastic Particle Methods
Speaker : 이선환 박사 (Lattice Engines Inc.) Date & Time : 2012. 2. 23. 목요일 오후 2:00 Where : 융대원 D-123 Abstract Particle methods, also known as Monte Carlo method in statistical community, becomes a powerful tool for a variety of research areas such as chemistry, astronomy and finance to list a few. This is mainly due to the enormous amount of advance in computational resource. In this seminar, more efficient and robust stochastic particle method using parallel methodology in general setting is introduced. The parallel methodology takes full advantage of Markov Chain random walk and its beautiful theories. The parallel stochastic particle method is applied to two different areas, optimal filtering problem and simulation of Integro-differential equation. In each application, the mathematical proof of convergence as well as the numerical example is provided. As the first application area, Markov Chain Distributed Particle Filter (MCDPF) algorithm is introduced and the performance of this method is demonstrated with a bearing only measurement target tracking numerical example and is further compared with an existing methods, Distributed Extended Kalman Filter (DEKF). As another application area, we applied parallel stochastic particle method to a stochastic simulation of large number of particle coagulation. Since more particles result in more accurate and reliable solution it is desirable to simulate this equation with more number of particles. By applying parallel stochastic particle method a comparable solution is obtained with multiple processors where one processor maintains much fewer particles while communicating to neighboring processors.
※자세한 내용은 첨부문서를 참고해 주시기 바랍니다. |
|||||
| 영상주소 | |||||
| 첨부파일 | 12.02.23_지융_이선환_박사.doc [75건 다운로드] |
||||
| 이전글 | 네팔 솔라 봉사단 뉴스레터 (2012.2.7) 2012-02-07 |
|---|---|
| 현재글 | 융대원 지능형융합시스템학과 세미나 안내 - 연사 : 이선환 박사 (Lattice Engines Inc.) 2012-02-20 |
| 다음글 | [입주기업 소식]나노브릭, 일본 나노테크에서 IT부문 최우수상...외국계 기업으로 이례적 수상 2012-02-21 |
