| Project/Area Number |
18079006
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research on Priority Areas
|
| Allocation Type | Single-year Grants |
|---|
| Review Section |
Science and Engineering
|
|---|
| Research Institution | Tokyo Institute of Technology |
|---|
Principal Investigator |
KABASHIMA Yoshiyuki Tokyo Institute of Technology, 大学院・総合理工学研究科, 教授 (80260652)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
渡辺 治 東京工業大学, 大学院・情報理工学研究科, 教授 (80158617)
竹田 晃人 東京工業大学, 大学院・総合理工学研究科, 助教 (70397040)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
WATANABE Osamu 東京工業大学, 大学院・情報理工学研究科, 教授 (80158617)
TAKEDA Koujin 東京工業大学, 大学院・総合理工学研究科, 助教 (70397040)
|
|---|
| Project Period (FY) |
2006 – 2009
|
| Project Status |
Completed (Fiscal Year 2009)
|
| Budget Amount *help |
¥44,800,000 (Direct Cost: ¥44,800,000)
Fiscal Year 2009: ¥12,400,000 (Direct Cost: ¥12,400,000)
Fiscal Year 2008: ¥14,100,000 (Direct Cost: ¥14,100,000)
Fiscal Year 2007: ¥14,100,000 (Direct Cost: ¥14,100,000)
Fiscal Year 2006: ¥4,200,000 (Direct Cost: ¥4,200,000)
|
|---|
| Keywords | 確率推論 / 確率伝搬法 / 要約伝搬法 / レプリカ対称性の破れ / スピングラス / レプリカ拡張 / レプリカ対称性 / 平均場近似 / 解析接続 / 複素零点評価 / レプリカ数 / レプリカ法 / ランダム行列理論 / TAP方程式 / 確立伝搬法 / 適応TPA法 / フィッシャー情報量 / ランダムネス / 理論計算機科学 |
|---|
| Research Abstract |
Approximate algorithms for computing probabilities, such as belief propagation, sometimes offer remarkable performance for "easy" inference problems that come out in information and communication. However, when the profile of the target distribution is of the "many-valley" structure, they do not necessarily lead to the good performance. In this research project, we studied a scheme for improving the inference performance by introducing the concept of replica symmetry breaking to the approximate algorithms, and examined its possibilities and limitations based on various case studies.
|
