Theoretical and Applied Data Science Seminar - Dr. Eric Price

Thursday, April 29, 2021 - 12:00pm to 1:00pm
Event Type: 

Title: Instance-Optimal Compressed Sensing via Conditional Resampling

Abstract: Given a distribution P of images, how many random linear measurements
are required for approximate recovery? Classical compressed sensing
gives an upper bound for approximately sparse P, but actual
distributions may have more or different structure than sparsity.  We
instead give an instance-optimal bound: when P is (approximately)
known, then _conditional resampling_, where we output a sample of
P(x | y), is within constant factors of the best possible recovery
algorithm.

When applied to state-of-the-art deep generative models, we find that
conditional resampling produces less washed-out, more realistic
results than prior methods.  It also has nice fairness properties,
including proportional representation: the representation of any group
in the output matches the input distribution.

 

Bio: Eric Price is an associate professor in the Department of
Computer Science at UT Austin, where he studies how algorithms can
produce more accurate results with less data.  He received a Ph.D. in
computer science from MIT in 2013. Eric's research was featured in
Technology Review's TR10 list of 10 breakthrough technologies of 2012,
his thesis received a George M. Sprowls award for best doctoral thesis
in computer science at MIT, and he has received an NSF CAREER
award. Two themes of his research are adaptivity, where initial data
can guide future data collection, and signal structure, where a
structural assumption can yield provable improvements in space or
sample complexity.

 

To join this Zoom Presentation, please click:

Join from a PC, Mac, iPad, iPhone or Android device:

    Please click this URL to start or join. https://iastate.zoom.us/j/97897667396?pwd=UHBYMSt4dmpkSGdMcVpwUnZIRS9Xdz09
    Or, go to https://iastate.zoom.us/join and enter meeting ID: 978 9766 7396 and password: TADSISU
 
Join from dial-in phone line:

    Dial: +1 312 626 6799 or +1 646 876 9923
    Meeting ID: 978 9766 7396
    Participant ID: Shown after joining the meeting
    International numbers available: https://iastate.zoom.us/u/abiHgDhQMS

After the presentation, there will be a short time for discussion and questions.