BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250626T233533Z
LOCATION:B302-B305
DTSTART;TZID=America/New_York:20241121T100000
DTEND;TZID=America/New_York:20241121T170000
UID:submissions.supercomputing.org_SC24_sess534_drs111@linklings.com
SUMMARY:Effects of Lossy Compression Data on Machine Learning Models
DESCRIPTION:Max Faykus (Clemson University)\n\nMachine learning is a funda
 mental tool that is incorporated in fields across academia and industry. D
 ue to the large amounts of data needed for training machine learning model
 s, compression is utilized because it reduces the data footprint playing a
  critical role in storage. Machine learning involves the use of algorithms
  and models to learn patterns in data allowing AI to make decisions withou
 t specific programming. On the other hand, compression utilizes encoding a
 nd decoding techniques to reduce file size. Compression can be lossy or lo
 ssless; lossy causes a loss of data while lossless preserves the data.\n\n
 This dissertation explores the accuracy and scalability of machine learnin
 g when working with lossy distorted data. Performance metrics studied look
  at how accurately the model’s inference performs. Issues with machine lea
 rning performance on lossy data involve the following: data storage, data 
 transfer bandwidth, and processing on the intersection between machine lea
 rning and lossy compression. Over these various issues, machine learning i
 s examined in different domains. This work investigates how meaningful pat
 terns in the distorted data are extracted.\n\nThe primary focus explores n
 eural network models' ability to manage lossy compressed data and find way
 s to mitigate loss due to distortion, addressing machine learning across v
 arious domains including object detection, semantic segmentation, and imag
 e classification, to find the balance between compression ratio and data q
 uality.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pass
 \n\nSession Chairs: Ayesha Afzal (Friedrich-Alexander University, Erlangen
 -Nuremberg; Erlangen National High Performance Computing Center); Sally El
 lingson (University of Kentucky); and Alan Sussman (University of Maryland
 )\n\n
END:VEVENT
END:VCALENDAR