馮興東，博士畢業(yè)于美國伊利諾伊大學香檳分校，現(xiàn)任上海財經(jīng)大學統(tǒng)計與管理學院院長、統(tǒng)計學教授、博士生導師。研究領域為數(shù)據(jù)降維、穩(wěn)健方法、分位數(shù)回歸以及在經(jīng)濟問題中的應用、大數(shù)據(jù)統(tǒng)計計算、強化學習等，在國際頂級統(tǒng)計學/計量經(jīng)濟學期刊JASA、AoS、JRSSB、Biometrika、JoE以及人工智能期刊/頂會JMLR、NeurIPS、ICML上發(fā)表論文多篇。2018年入選國際統(tǒng)計學會推選會員(Elected member)，2019年擔任全國統(tǒng)計教材編審委員會第七屆委員會專業(yè)委員（數(shù)據(jù)科學與大數(shù)據(jù)應用組），2020年擔任第八屆國務院學科評議組（統(tǒng)計學）成員，2022年擔任全國應用統(tǒng)計專業(yè)碩士教指委委員，2023年擔任全國工業(yè)統(tǒng)計學教學研究會副會長以及中國數(shù)學會概率統(tǒng)計分會常務理事，2022和2023年起分別兼任國際統(tǒng)計學權(quán)威期刊Annals of Applied Statistics和Statistica Sinica編委（Associate Editor）以及國內(nèi)統(tǒng)計學權(quán)威期刊《統(tǒng)計研究》編委。
報告摘要：This paper presents a generalization study of kernel quantile regression with random features (KQR-RF), which accounts for the non-smoothness of the check loss in KQR-RF by introducing a refined error decomposition and establishing a novel connection between KQR-RF and KRR-RF. Our study establishes the capacity-dependent learning rates for KQR-RF under mild conditions on the number of RFs, which are minimax optimal up to some logarithmic factors. Importantly, our theoretical results, utilizing a data-dependent sampling strategy, can be extended to cover the agnostic setting where the target quantile function may not precisely align with the assumed kernel space. By slightly modifying our assumptions, the capacity-dependent error analysis can also be applied to cases with Lipschitz continuous losses, enabling broader applications in the machine learning community. To validate our theoretical findings, simulated experiments and a real data application are conducted.
