Distributed Algorithms for Principal Component Analysis by Dr. Anna Scaglione
Many data analysis algorithm in array processing and machine learning rely on Principal Component Analysis (PCA) as basic primitive. Distributed PCA algorithms are useful when the data are acquired by a network of sensors, or when the data are split over several servers due to storage limitations. Such algorithms can harness local communications and network connectivity to overcome the need of communicating and accessing the entire array locally, but the way the different datasets are divided across the network informs the algorithm’s design and affects its performance limits. The talk will explain how one can decompose the problem of computing the principal components in general, by classifying the methodologies to perform distributed PCA on different datasets, and discussing their performance and their applications in the context of distributed data acquisition systems.
Date and Time
Location
Hosts
Registration
- Date: 16 Jul 2019
- Time: 06:00 PM to 08:00 PM
- All times are (GMT-06:00) US/Central
-
Add Event to Calendar
- Avenida Patria 1201
- Colonia Lomas del Valle
- Zapopan, Jalisco
- Mexico 45129
- Building: Sala Neuberger
- Contact Event Host
- Co-sponsored by Dr. Gilberto Ochoa
- Survey: Fill out the survey
- Starts 01 April 2019 10:00 AM
- Ends 16 July 2019 10:09 PM
- All times are (GMT-06:00) US/Central
- No Admission Charge
Speakers
Dr. Scaglione of Arizona State University
Distributed Algorithms for Principal Component Analysis
Biography:
Anna Scaglione (M.Sc.'95, Ph.D. '99) is currently a professor in electrical and computer engineering at Arizona State University. She was Professor of Electrical Engineering previously at the at UC Davis (2010-2014), Associate Professor at UCD 2008-2010 and at Cornell (2006-2008), and Assistant Professor at Cornell (2001-2006) and at the University of New Mexico (2000-2001).
Her expertise is in the broad area of statistical signal processing for communication, electric power systems and networks. Her current research focuses on studying and enabling decentralized learning and signal processing in networks of sensors.
Dr. Scaglione was elected an IEEE fellow in 2011. She served as Associate Editor for the IEEE Transactions on Wireless Communications and on Signal Processing, as EiC of the IEEE Signal Processing letters. She was member of the Signal Processing Society Board of Governors from 2011 to 2014. She received the 2000 IEEE Signal Processing Transactions Best Paper Award and more recently was honored for the 2013, IEEE Donald G. Fink Prize Paper Award for the best review paper in that year in the IEEE publications, her work with her student earned 2013 IEEE Signal Processing Society Young Author Best Paper Award (Lin Li).
Email:
Address:Arizona State University, Office: GWC 310 Lab: GWC 325, Tempe, United States, 85287-9309