学　术

　　为搭建我校博士后之间的学术交流平台，促进学术水平提升，学校博士后管理办公室组织开展博士后学术沙龙活动。本次沙龙由我校博士后张秋燕、丁可伟、王中宝分享其研究成果，诚挚邀请感兴趣的师生参加。

　　一、时　间：2018年11月20日（周二）14：00

　　二、地　点：清水河校区　经管楼宾诺咖啡

　　三、活动安排：

　　报告一         

　　（1）主题：Local bifurcations of the enzyme-catalyzed reaction comprising a branched network

　　（2）主讲人：张秋燕  数学科学学院博士后

　　（3）交流内容：

　　An enzyme-catalyzed reaction system with four parameters a, b, c, κ is discussed. The system　can be reduced to a quartic polynomial differential system with four parameters, which leads　to difficulties in the computation of semi-algebraic systems of large degree polynomials. Those　systems have to be discussed on subsets of special biological sense, none of which is closed under operations of the polynomial ring. In this paper, we overcome those difficulties to determine the exact number of equilibria and their qualitative properties. Moreover, we obtain parameter conditions for all codimension-1 bifurcations such as saddle-node, transcritical, pitchfork and Hopf bifurcations. We compute varieties of Lyapunov quantities under the limitations of biological requirements and prove that the weak focus is of at most order 2. We further obtain parameter conditions for exact number of limit cycles arising from Hopf bifurcations. Moreover, we give conditions for the existence of a cusp and compute the parameter curves for the Bogdanov-Takens bifurcation, which induces the appearance of homoclinic orbits and periodic orbits, indicating the tendency to steady-states or a rise of periodic oscillations for the concentrations of the substrate and the product.

　　（4）主讲人简介：

　　Qiuyan Zhang is a post-doc majored in differential equations and dynamic systems. She obtained PhD in 2016, advised by Prof.Weinian Zhang in Sichuan University. Her main research interests include stability discussion, center determination and periodic orbits’ exsistence of high dimensional degenerate vector field and applications in biochemical reactions, complex Networks and mechanical oscillation.     

　　报告二

　　（1）主题：Distributionally robust chance constrained problems under general moments information

　　（2）主讲人：丁可伟  数学科学学院博士后

　　（3）交流内容：

　　Distributionally robust chance constrained programming (DRCCP)  has become a significant and effective approach to address practical optimization problems involving uncertainty. Most distributionally robust chance constrained methods are developed with the purpose of achieving a computationally tractable model for corresponding DRCCPs under the exact second order moment information. But it is not reliable when the empirical moments has been treated as the exact moments. In order to strengthen the safety, we focus on DRCCPs under general moments information. By convex analysis, we obtain the equivalent convex programming form for DRCCP under assumptions that the first and second order moments belong to the convex and compact set. We give some examples of support functions about matrix sets to show the tractability of the equivalent convex programming and obtain the closed form solution for the worst case VaR optimization problem. Then, we present the equivalent convex programming form for DRCCP under assumptions that the set of first order moment and the support subsets are convex and compact, and the equivalent form for distributionally robust nonlinear chance constrained problem under the set of first order moment and the convex support set are convex and compact. We also show some tractability examples.

　　（4）主讲人简介：

　　Kewei Ding is a post-doc majored in stochastic optimization. He obtained bachelor’s degree in mathematics and applied mathematics in Sichuan University, then performed PhD research in financial mathematics and econometrics advised by Prof. Nanjing Huang in Sichuan University. His main research interests include robust optimization; distributionally robust optimization; chance constrained programming and portfolio management.

　　报告三

　　（1）主题：A new projection-type method for solving multi-valued mixed variational inequalities without monotonicity

　　（2）主讲人：王中宝  数学科学学院博士后

　　（3）交流内容：

　　Multi-valued mixed variational inequalities (MMVI) is an important extension of variational inequalities and have wide applications in market economy and electrical circuits. One of important problems associated with MMVI is the development of efficient iterative algorithms to compute approximate solutions. Because of the appearance of the nonlinear term and single-valued mapping replaced by multi-valued ones, constructing some effective methods for MMVI is more difficult than for mixed variational inequalities. Therefore, it is worth developing methods for solving MMVI.  In this talk, we will present a new projection-type algorithm for solving multi-valued mixed variational inequalities without monotonicity. Under some suitable assumptions, we will show that the sequence generated by the proposed algorithm converges globally to a solution of the multi-valued mixed variational inequality considered. Our algorithm is based on the generalized f-projection operator rather than the well-known resolvent operator. Preliminary computational experience is also reported.

　　（4）主讲人简介：

　　Zhong-bao Wang is a post-doc majored in variational inequalities with applications. He obtained Master’s degree in Operational Research and Cybernetics advised by Prof. Xie-ping Ding in Sichuan Normal University, then performed PhD research in Operational Research and Cybernetics advised by Prof. Nanjing Huang in Sichuan University.

　　三、主办单位：电子科技大学博士后管理办公室

　　四、承办单位：数学科学学院

　　　　　　　　　电子科技大学博士后联谊会

　　　　　　　　　　　　　　　　　　　　电子科技大学博士后管理办公室

　　　　　　　　　　　　　　　　　　　　　　　2018年11月20日