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:20260924T172451Z
LOCATION:B304
DTSTART;TZID=America/New_York:20241118T112500
DTEND;TZID=America/New_York:20241118T114000
UID:submissions.supercomputing.org_SC24_sess749_ws_drbsd111@linklings.com
SUMMARY:Enhancing Lossy Compression Through Cross-Field Information for Sc
 ientific Applications
DESCRIPTION:Youyuan Liu (Temple University), Wenqi Jia (University of Texa
 s at Arlington), Taolue Yang (Temple University), Miao Yin (University of 
 Texas at Arlington), and Sian Jin (Temple University)\n\nLossy compression
  is one of the most effective methods for reducing the size of scientific 
 data containing multiple data fields. It reduces information density throu
 gh prediction or transformation techniques to compress the data. Previous 
 approaches use local information from a single target field when predictin
 g target data points, limiting their potential to achieve higher compressi
 on ratios. In this paper, we identified significant cross-field correlatio
 ns within scientific datasets. We propose a novel hybrid prediction model 
 that utilizes CNN to extract cross-field information and combine it with e
 xisting local field information. Our solution enhances the prediction accu
 racy of lossy compressors, leading to improved compression ratios without 
 compromising data quality. We evaluate our solution on three scientific da
 tasets, demonstrating its ability to improve compression ratios by up to 2
 5% under specific error bounds. Additionally, our solution preserves more 
 data details and reduces artifacts compared to baseline approaches.\n\nTag
 : Data Compression, Data Movement and Memory, Middleware and System Softwa
 re\n\nRegistration Category: Workshop Reg Pass\n\nSession Chairs: Sheng Di
  (Argonne National Laboratory (ANL), University of Chicago); Ana Gainaru (
 Los Alamos National Laboratory (LANL)); Sian Jin (Temple University); Xin 
 Liang (Oregon State University); Kento Sato (RIKEN Center for Computationa
 l Science (R-CCS)); and Dingwen Tao (Institute of Computing Technology (IC
 T), Chinese Academy of Sciences (CAS); University of Chinese Academy of Sc
 iences)\n\n
END:VEVENT
END:VCALENDAR
