Dr. Lei Li
Electrical Engineering and Computer Science Department
University of California, Berkeley
Time series data arise in numerous applications, such as motion capture, computer network monitoring, data center monitoring, environmental monitoring and many more. Finding patterns and learning features in such collections of sequences are crucial to solve real-world, domain specific problems, for example, to build humanoid robots, to detect pollution in drinking water, and to identify intrusion in computer networks.
In this talk, we focus on fast algorithms on mining co-evolving time series, with or without missing values. We will present a series of our effort in analyzing those data: (a) time series mining and summarization with missing values, and (b) learning features from multiple sequences. Algorithms proposed in the first work allow us to obtain meaningful patterns effectively and efficiently, and enable further mining tasks including forecast, compression, and segmentation for co-evolving time series, even with missing values. We also present “PLiF” and “CLDS”, novel algorithms to extract features from multiple sequences. Such learned features will can be exploited in many applications for time series such clustering and similarity search. In addition, we will briefly mention several other problems and algorithms, including natural motion stitching, bone constrained occlusion filling, a parallelization of our algorithms for multi-core systems, and algorithms for forecasting datacenter thermal condistions with complex cyber-physical interactions.
Dr. Lei Li is a Post-Doctoral researcher at EECS department of UC Berkeley and visiting researcher at CMU. His research interest lies in the intersection of machine learning, statistical inference and database systems. Specifically, he has been working on Bayesian inference in open universe probabilistic models, probabilistic programming language, large-scale learning, time series, communication and social networks. He has served in the Program Committee for SDM 2013, IJCAI 2011/2013, ICDM 2011 workshop on collective intelligence, KDD 2011, 2012 workshop on Multimedia data mining. He has been invited as reviewer for TOMCCAP,DAMI, KDD, SIGMOD,VLDB,PKDD, and WWW.
Lei received his B.S. in Computer Science and Engineering from Shanghai Jiao Tong University in 2006 and Ph.D. in Computer Science from Carnegie Mellon University in 2011, respectively. His dissertation work on fast algorithms for mining co-evolving time series was awarded ACM KDD best dissertation (runner up).