CS 201: Approximate Nearest Neighbor Methods to Rapidly Search Large Seismic Signal Archives, ANTONIO GONZALES, Sandia National Laboratories

Speaker: Antonio Gonzales
Affiliation: Sandia National Laboratories

ABSTRACT: Waveform correlation is a proven method for identifying the origin of seismic signals. Unfortunately, waveform correlation is computationally expensive which either limits the size of the waveform template archive or requires access to large distributed systems.  Recently, Sandia has started research into using approximate nearest neighbors in kernel space to greatly increase the efficiency of finding similar waveforms in large databases.  This talk will provide a high-level overview of the method and present some early results of this line of research. BIO: Antonio I. Gonzales is a Principal Member of the Technical staff at Sandia National Laboratories (SNL) in Albuquerque, New Mexico. He serves as the lead for a team that solves complex signal processing problems and has worked for nearly 15 years in the domains of signal processing, machine learning, and software architectures. Antonio received his Bachelors of Science (2002) from the University of Oklahoma and his Masters of Computer Science (2003) from the University of Illinois at Urbana-Champaign.

Hosted by Professor Peter Reiher

Date/Time:
Date(s) - Oct 17, 2017
4:15 pm - 5:45 pm

Location:
3400 Boelter Hall
420 Westwood Plaza Los Angeles California 90095