Invited Speaker

Junchi Yan, Shanghai Jiao Tong University, China
严骏驰，上海交通大学

严骏驰，上海交通大学计算机系长聘轨副教授，人工智能重点实验室主任助理。科技部2030新一代人工智能青年科学家项目负责人、自然基金委优青、教育部人工智能资源建设深度学习首席专家。曾任IBM中国研究院首席研究员。获中国计算机学会优博、人工智能学会吴文俊优青等科技奖励。主要研究方向为机器学习，特别是与运筹优化、量子计算等领域的结合。发表CCF-A类第一/通讯作者论文近百篇，授权美国发明专利超30项，谷歌学术引用超过8000次。任ICML、NeurIPS、CVPR、AAAI等会议领域主席、Pattern Recognition期刊编委。
Dr. Junchi Yan is currently an Associate Professor with Department of Computer Science and Engineering, Shanghai Jiao Tong University. He is currently the PIs of the MOST 2030 Next AI Major Project and the NSFC Outstanding Youth Fund, and the Chief Expert for MOE curricula development in the Deep Learning Area. He was once the (Principal) Researcher with IBM Research from 2011-2018. He was the awardee of CCF Outstanding PhD Thesis. His main research interests include machine learning and its intersection with quantum computing and operational research. He has published nearly 100 CCF-A papers as first/correspondence authors and have 30+ authorized US patents, with Google Scholar citations over 8000. He regularly served as AC for ICML/NeurIPS/CVPR etc. and is on the editorial board of Pattern Recognition Journal.

报告题目：图论问题的机器学习求解
图论与组合问题多具有高度计算复杂度，传统算法多依赖专家人工设计。而数据驱动的方式对复杂问题进行求解，则成为近年来的一个新兴方向。本报告将分享课题组近年来在基于机器学习的图论与组合问题求解上的相关理论与算法的研究工作，及在EDA、联邦学习等场景的应用探索。最后，也将简要介绍项目组在量子人工智能，特别是量子图机器学习方面的最新进展。相关工作主要发表在TPAMI、NeurIPS、ICML、SIGKDD等机器学习期刊和会议。

Speech Title: Machine Learning for Graph and Combinatorial Problems and Beyond
Graph and combinatorial problems often incur inherent high complexity, and the traditional solvers are mostly based on human expert design which can be costive and challenging. On the other hand, data-driven approaches like machine learning, especially deep learning, have shown promising impact in many perception tasks while their role in the above problems are relatively in its early stage. In this talk, I will share our recent progress in the area of machine learning for solving combinatorial problems on graphs, and their applications in EDA, federated learning, etc. I will also briefly give some initial results on quantum graph learning. The results o f this talk have been published in TPAMI、NeurIPS、ICML、ICLR, SIGKDD, CVPR.