報告人: 夏志明教授、薄立軍教授、許勇教授
講座日期:2020-10-23
講座時間:8:30
報告地點:騰訊會議(404445121)、數(shù)學與信息科學學院學術(shù)交流廳
主辦單位:數(shù)學與信息科學學院
報告題目1:Distributed Feature Screening via Componentwise Debiasing
報告人: 夏志明教授
講座日期:2020-10-23
講座時間:8:30
報告地點:騰訊會議(404445121)
講座人簡介:
夏志明,教授,博士生導師,西北大學現(xiàn)代統(tǒng)計研究中心副主任,統(tǒng)計系主任,主要致力于張量數(shù)據(jù)分析、大數(shù)據(jù)異質(zhì)性結(jié)構(gòu)推斷、分布式統(tǒng)計推斷與計算、生物統(tǒng)計學等數(shù)據(jù)科學理論與應用研究。在“Biometrika”、“Journal of machine learning research”, “Technometrics”、“Statistics in Medicine”等國際統(tǒng)計與機器學習期刊以及“中國科學”、“應用概率統(tǒng)計”等國內(nèi)期刊發(fā)表論文40余篇;主持國家自然科學基金項目3項,主持省部級項目3項, 作為骨干成員獲得“陜西省科學技術(shù)進步獎”二、三等獎共2項,“陜西省高??茖W技術(shù)獎”一等獎共2項,“陜西省國防科技進步獎”一等獎1項;先后赴香港科技大學、佛羅里達大學等科研機構(gòu)進行專業(yè)訪問與學術(shù)交流。
講座簡介:
Feature screening is a powerful tool in processing high-dimensional data. When the sample size N and the number of features p are both large, the implementation of classic screening methods can be numerically challenging. In this paper, we propose a distributed screening framework for big data setup. In the spirit of “divide-and-conquer”, the proposed framework expresses a correlation measure as a function of several component parameters, each of which can be distributively estimated using a natural U-statistic from data segments. With the component estimates aggregated, we obtain final correlation estimate that can be readily used for screening features. This framework enables distributed storage and parallel computing and thus is computationally attractive. Due to the unbiased distributive estimation of the component parameters, the nal aggregated estimate achieves a high accuracy that is insensitive to the number of data segments m. Under mild conditions, we show that the aggregated correlation estimator is as ecient as the centralized estimator in terms of the probability convergence bound and the mean squared error rate; the corresponding screening procedure enjoys sure screening property for a wide range of correlation measures. The promising performances of the new method are
supported by extensive numerical examples.
報告題目2:Optimal Tracking Portfolio with A Ratcheting Capital Benchmark
報告人: 薄立軍教授
講座日期:2020-10-23
講座時間:14:30
報告地點:數(shù)學與信息科學學院學術(shù)交流廳
講座人簡介:
薄立軍,西安電子科技大學數(shù)學與統(tǒng)計學院教授,本科畢業(yè)于西安電子科技大學數(shù)學系,碩士和博士畢業(yè)于南開大學概率論與數(shù)理統(tǒng)計專業(yè),研究方向為隨機分析、隨機控制與金融數(shù)學。 2012年入選教育部新世紀優(yōu)秀人才支持計劃,先后主持國家自然科學基金面上項目2項 (數(shù)理學部)、中科院前沿科學重點研究計劃-青年拔尖科學家項目。目前已在國際公認的概率統(tǒng)計、金融數(shù)學、管理和運籌學權(quán)威期刊Math. Finan., Finan. & Stoch., SIAM J. Finan. Math., SIAM J. Control & Optim, Math. Opers. Res., J.Banking & Finan., Appl. Math. & Optim., J. Dyn. Econ. & Contr., Queueing Syst., J. Theor. Probab.上發(fā)表學術(shù)論文30余篇。目前擔任中國概率統(tǒng)計學會會刊《應用概率統(tǒng)計》編委;美國數(shù)學科學研究所(AIMS)旗艦期刊《J. Dynamics & Games》Associate Editor.
講座簡介:
This talk is concerned with the ?nite horizon portfolio management by optimally tracking a ratcheting capital benchmark process. To formulate such an optimal tracking problem, we envision that the fund manager can dynamically inject capital into the portfolio account such that the total capital dominates the nondecreasing benchmark ?oor process at each intermediate time. The control problem is to minimize the cost of the accumulative capital injection. We ?rst transform the original problem with ?oor constraints into an unconstrained control problem, however, under a running maximum cost. By identifying a controlled state process with re?ection, we next transform the problem further into an equivalent auxiliary problem, which leads to a nonlinear Hamilton-Jacobi-Bellman (HJB) with a Neumann boundary condition. By employing the dual transform, the probabilistic representation approach and some stochastic ?ow arguments, the existence of the unique classical solution to the dual HJB is established. The veri?cation theorem is carefully proved, which gives the complete characterization of the primal value function and the feedback optimal portfolio.
報告題目3:非高斯Lévy噪聲誘導的復雜動力學研究
報告人: 許勇教授
講座日期:2020-10-23
講座時間:16:30
報告地點:數(shù)學與信息科學學院學術(shù)交流廳
講座人簡介:
許勇,博士,教授,博導,德國洪堡高級研究學者(experienced researcher),教育部新世紀優(yōu)秀人才,陜西省青年科技新星,陜西省首屆杰出青年基金獲得者,陜西省特支計劃科技創(chuàng)新人才,西北工業(yè)大學優(yōu)秀青年教師、翱翔青年學者。主要研究領域為應用數(shù)學和力學,研究方向為隨機動力系統(tǒng)與應用概率統(tǒng)計。主持包括6項國家自然科學基金在內(nèi)的10余項省部級項目,在國內(nèi)外著名期刊上發(fā)表 SCI期刊論文100余篇,其中第一或通訊作者80余篇。研究成果分別以第一完成人獲得2016年度陜西省科學技術(shù)一等獎和2016年教育部自然科學技術(shù)二等獎,2007年獲得陜西省科學技術(shù)一等獎(第四完成人),2018年陜西省第十一屆青年科技獎獲得者。2019年陜西省重點科技創(chuàng)團隊負責人。現(xiàn)任陜西省振動工程學會理事長、中國力學學會動力學與控制專業(yè)委員會隨機動力學專業(yè)組組長、中國數(shù)學會理事、中國振動工程學會理事、陜西省工業(yè)與應用數(shù)學學會副理事、陜西省統(tǒng)計學會常務理事、中國振動工程學會非線性振動專業(yè)委員會委員、中國宇航學會飛行器任務規(guī)劃專業(yè)委員會委員, 國際雜志Complexity與Frontiers in Physiology/Physics/Molecular Biosciences-Biophysics副主編(Associate Editor),Chaos,Theoretical & Applied Mechanics Letters,動力學與控制學報編委、美國《數(shù)學評論》評論員、60余本國內(nèi)外期刊審稿人。教學上主要從事隨機動力系統(tǒng)、統(tǒng)計計算、近現(xiàn)代統(tǒng)計、隨機微分方程、非線性動力學等本科、研究生和留學生課程的教學,編著教材4本。第一完成人獲得陜西省教學成果特等獎1項。
講座簡介:
報告將給出作者及其團隊在非高斯Levy噪聲所誘導的復雜動力學方面的成果,包括早期預警,粗糙勢函數(shù)與隨機動力學,隨機二元機翼模型,隨機平均原理等,特別是發(fā)現(xiàn)以往高斯噪聲下所不能發(fā)現(xiàn)的現(xiàn)象,研究非高斯與相關(guān)噪聲的對非線性動力學的作用機理、方法和理論。最后給出目前存在的問題以及可能解決的方法。