Event Date
SPEAKER: John Wu; Lawrence Berkeley National Lab
TITLE: “Statistical Similarity for Data Compression”
ABSTRACT: Scientific data is usually collected to a greater degree of precision than is significant, which produces random-looking data that are hard to compress with established techniques. We propose a class of lossy compression techniques based on locally exchangeable measure that captures the distribution of repeating data blocks while preserving unique patterns. The technique has been demonstrated to reduce data volume by more than 100-fold on power grid monitoring data where a large number of data blocks can be characterized as following stationary probability distributions. In this talk, we will review a number of recent developments that allows us to handle non-stationary time series, multidimensional time series, and data with cyclic distributions.
SPEAKER BIO: Dr. Wu works actively on a number of topics in data management, data analysis, and high-performance computing. His algorithmic research work includes statistical methods for feature extraction, indexing techniques for searching large datasets, and matrix based techniques for machine learning and scientific computing. He is the developer of a number of software packages, including, IDEALEM, SDS, FastBit and TRLan. Among them, the FastBit indexing software has won R\&D 100 Award, and is used by many organizations. For example, a German bioinformatics company uses FastBit to accelerate their molecular docking software by hundreds of times, and an US internet company uses it to sift through terabytes of advertisement related data daily. A FastBit paper is counted among the 40 major works funded by DOE Office of Science, as a part of its 40th Anniversary celebration in 2018.
DATE: Thursday, April 4th, 4:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm MSB 4110 (4th floor lounge)
STA 290 Seminar List: https://statistics.ucdavis.edu/seminars