We introduce a stochastic interacting particle consensus system for global optimization of high dimensional non-convex functions. This algorithm does not use gradient of the function thus is suitable for non-smooth functions. We prove, for fully discrete systems, that under dimension-independent conditions on the parameters, with suitable initial data, the algorithms converge to the neighborhood of the global minimum almost surely. We also introduce an Adaptive Moment Estimation (ADAM) based version to significantly improve its performance in high-space dimension.
About the Lecturer:
Jin Shi, President and Chair Professor of Shanghai Jiao Tong University, Institute of Natural Sciences; Director of Shanghai National Center for Applied Mathematics, Director of Shanghai Jiao Tong University, Key Laboratory of Scientific and Engineering Computing, Ministry of Education and Director of Center for Artificial Intelligence Mathematics; awarded with Feng Kang Prize for Scientific Computing, shortlisted in National Talents Program; elected as AMS Fellow, SIAM Fellow, Foreign Member of Academia Europaea-Academy of Europe, Fellow of European Academy of Sciences, etc.
Rui Hongxing, Professor from School of Mathematics
10:00, November 11 (Thursday)
Tencent Meeting, ID: 940 638 639, password: 211111
Hosted by: School of Mathematics, Shandong University