报告标题: Tensor Train Factorization with Temporal Smoothness for Low-Rank Tensor Completion with Applications
报告人：喻高航 教授
报告时间：2022/10/14（周五），16:30-17:30
报告地点：南山1-415
报告摘要: Estimating data from an incomplete measurement or observation plays an important role in the area of big data analytic, visual data recovery and network engineering, etc. In this talk, we present two Tensor-Train-factorization-based methods for low rank tensor completion problem. In order to get better performance for real-world data completion, the proposed models make full use of the potential periodicity and inherent correlation properties via temporal regularization. Extensive numerical experiments on internet traffic data, color images, and gray scale video show that the proposed model outperforms many existing state-of-the-art methods.
报告人简介：喻高航，杭州电子科技大学教授、博导，主要从事张量数据分析、大规模优化计算及其在机器学习、图像处理与医学影像中的应用研究。先后在SIAM Journal on Imaging Sciences, International Journal of Robust and Nonlinear Control，IEEE Signal Processing Letters，Journal of Mathematical Imaging and Vision等国际期刊上发表40余篇SCI论文,先后主持5项国家自然科学基金、1项教育部新世纪优秀人才支持计划项目和1项浙江省自然科学基金重大项目，有多篇论文入选ESI高被引榜单。自2013年起任国际学术期刊Statistics, Optimization and Information Computing执行编委（Coordinating Editor）。
