報(bào)告人：喻高航 教授
報(bào)告題目：Large Scale Tensor Decomposition: Randomized method and Its Applications
報(bào)告時(shí)間：2025年11月26日（周三）上午11：00
報(bào)告地點(diǎn)：云龍校區(qū)6號(hào)樓304會(huì)議室
主辦單位：數(shù)學(xué)與統(tǒng)計(jì)學(xué)院、數(shù)學(xué)研究院、科學(xué)技術(shù)研究院
報(bào)告人簡(jiǎn)介：
喻高航，浙江科技大學(xué)教授、博導(dǎo)，主要從事張量數(shù)據(jù)分析、大規(guī)模優(yōu)化計(jì)算及其在機(jī)器學(xué)習(xí)、圖像處理與醫(yī)學(xué)影像中的應(yīng)用研究。先后在SIAM Journal on Imaging Sciences, IEEE Transactions on Computational Social Systems，Expert Systems with Applications，Knowledge-Based Systems，Journal of Scientific Computing，Applied Mathematical Modelling,Inverse Problems, Journal of Optimization Theory and Applications, Optimization Methods and Software等國(guó)際期刊上發(fā)表50余篇SCI論文,先后主持5項(xiàng)國(guó)家自然科學(xué)基金、1項(xiàng)教育部新世紀(jì)優(yōu)秀人才支持計(jì)劃項(xiàng)目和1項(xiàng)浙江省自然科學(xué)基金重大項(xiàng)目，有多篇論文入選ESI高被引榜單。現(xiàn)任國(guó)際SCI學(xué)術(shù)期刊Intelligent Automation & Soft Computing 的期刊編委；國(guó)際學(xué)術(shù)期刊Statistics, Optimization and Information Computing執(zhí)行編委（Coordinating Editor）。
報(bào)告摘要：
Low-rank approximation of tensors has been widely used in high-dimensional data analysis. It usually involves singular value decomposition (SVD) of large-scale matrices with high computational complexity. Sketching is an effective data compression and dimensionality reduction technique applied to the low-rank approximation of large matrices. This talk presents some efficient randomized algorithms for low-rank tensor approximation based on T-product, Tucker decomposition, with rigorous error-bound analysis. We also present some applications on tensor completion and parameter-efficient-fine-tuning (PEFT) for transfer learning.