Natural language research announcements

[Julia Hockenmaier <juliahmr AT illinois.edu>]

Begin forwarded message:

From: "Schaefer, Melinda M" <mschaefr AT illinois.edu>

Date: April 27, 2010 3:27:52 PM CDT

To: "faculty AT cs.illinois.edu" <faculty AT cs.illinois.edu>, "cs-grads AT cs.uiuc.edu" <cs-grads AT cs.uiuc.edu>, "postdocs AT cs.uiuc.edu" <postdocs AT cs.uiuc.edu>, "announce AT cs.uiuc.edu" <announce AT cs.uiuc.edu>

Cc: "clerical AT cs.uiuc.edu" <clerical AT cs.uiuc.edu>

Subject: Seminar: Models for Mixed Signals - MONDAY, May 3, 2PM

 

University of Illinois at Urbana-Champaign

Department of Computer Science

Electrical and Computer Engineering

 

Models for Mixed Signals

 

 

Paris Smaragdis

Senior Research Scientist

Adobe Systems

 

Monday, May 3, 2010, 2:00 p.m.*

2405 Thomas M. Siebel Center for Computer Science

201 North Goodwin Avenue

Urbana, Illinois 61801-2302 USA

 

 

Abstract:

 

Dealing with superimposed signals is one of the most challenging problems we face today in audio and speech processing.  In this talk I'll present a new way to approach this problem and some state-of-the-art methods for working with mixed signals.  When suitably abstracted, mixed signals exhibit a unique geometry which points to a combinatorially hard solution.  Using probabilistic modeling we can derive a family of efficient sparse learning approaches that obtain superior results in a variety of mixed signal tasks.  In this talk I'll present some of these algorithms and how they found their way to real-life applications.

 

 

Biography:

 

Paris Smaragdis is a senior research scientist at Adobe Systems.  He completed his graduate and postdoctoral studies at MIT, where he conducted research on computational perception.  Prior to Adobe he was a research scientist at Mitsubishi Electric Research Labs, during which time he was selected by the MIT Technology Review as one of the top 35 young innovators of 2006.  Paris' research interests lie in the intersection of machine learning and signal processing.

 

 

*A reception will be held outside of 2405 SC immediately following the seminar

 

Attachment: Paris_Poster.pdf
Description: Adobe PDF document