报告人：Han Mao Kiah
报告题目：Low Bandwidth Algorithms for Reed-Solomon Coded Data
时间：2025年6月25日15:00-15:30
地点：数学楼423
摘要：
In this talk, we explore methods to address two distributed computing applications with low bandwidth requirements.For the first part of the talk, we examine the repair problem in distributed storage systems. Specifically, we revisit the Guruswami-Wootters repair scheme for Reed-Solomon codes and ask: Can we correctly repair a failed node in the presence of erroneous nodes? Equivalently, we study the downloaded traces as a code and investigate their code-distance properties.In the second part, we study the problem of multivariate polynomial evaluation in a distributed computing system comprising a single master node and multiple worker nodes. Specifically, we generalize the Lagrange Coded Computing framework (Yu et al., 2019), proposing a scheme that enables simultaneous computation while ensuring robustness against stragglers, erroneous workers, and colluding workers. The proposed scheme introduces a modest computational overhead, but significantly reduces download costs and provides straggler resistance comparable to existing solutions.
报告人简介：
Han Mao Kiah received the Ph.D. degree in mathematics from Nanyang Technological University (NTU), Singapore, in 2014. From 2014 to 2015, he was a Post-Doctoral Research Associate with the Coordinated Science Laboratory, University of Illinois at Urbana-Champaign. From 2015 to 2018, he was a Lecturer with the School of Physical and Mathematical Sciences (SPMS), NTU, where he is currently an Assistant Professor. His research interests include DNA-based data storage, coding theory, enumerative combinatorics, and combinatorial design theory.
邀请人：魏恒嘉 教授
