學(xué)術(shù)報(bào)告:A unified approach to variable selection for the non-mixture transformation cure model with interval-censored data and application to Alzheimer's disease
發(fā)布時(shí)間:2026-03-13
報(bào) 告 題 目 : A unified approach to variable selection for the non-mixture transformation cure model with interval-censored data and application to Alzheimer's disease
報(bào)告人:趙波,長(zhǎng)春工業(yè)大學(xué) 博士
報(bào)告時(shí)間:2026 年 3 月 15 日,15:30
邀請(qǐng)人:陳鋒
報(bào)告地點(diǎn):團(tuán)風(fēng)報(bào)告廳、#騰訊會(huì)議:804-620-1330
Abstract : In this talk, we study three-dimensional quasi-static electroporoelasticity equations to establish a well-posedness theory and propose some numerical approximations. The electroporoelasticity model, which comprises of Maxwell's equations with Biot's equations, plays a critical role in many fields such as oil-gas exploration, water conservancy exploration, earthquake early warning and some other areas in geophysics, etc. We first apply a Galerkin method to set up the well-posedness. Then, we study a splitting finite element approximation to improve the computational efficiency. In view of distinct time scales of the development of electromagnetic waves and elastic waves, we construct a multi-time stepping numerical algorithm to accelerate numerical computations. We also investigate the numerical stability as the Lamé constant tending to infinity or the hydraulic conductivity varying randomly. Finally, we present some numerical experiments to validate all theoretical findings.
報(bào)告人簡(jiǎn)介:趙波,理學(xué)博士,主要研究領(lǐng)域?yàn)樯锝y(tǒng)計(jì)復(fù)雜區(qū)間刪失數(shù)據(jù) 聯(lián)合建模,半?yún)?shù)統(tǒng)計(jì)推斷方法。 在 Analysis and Applications,Biometrical Journal , Statistical Papers , Lifetime Data Analysis , International Journal of Biostatistics 等期刊發(fā)表 SCI 檢索論文 12 篇。主持吉林省自然科學(xué)基金項(xiàng)目 1 項(xiàng),吉林省博士后擇優(yōu)資助項(xiàng)目 1 項(xiàng),獲吉林省自然科學(xué)成果獎(jiǎng)三等獎(jiǎng) 1 項(xiàng)。























