BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260522T150115Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess322_post236@linklings.com
SUMMARY:MATEDOR: MAtrix, TEnsor, and Deep-Learning Optimized Routines
DESCRIPTION:Ahmad Abdelfattah, Jack Dongarra, Stanimire Tomov, and Ichitar
 o Yamazaki (University of Tennessee) and Azzam Haidar (Nvidia Corporation)
 \n\nThe MAtrix, TEnsor, and Deep-learning Optimized Routines (MATEDOR) pro
 ject develops software technologies and standard APIs, along with a sustai
 nable and portable library, for large-scale computations that can be broke
 n down into very small matrix or tensor computations. The main target of M
 ATEDOR is to accelerate applications from important fields that fit this p
 rofile, including deep learning, data mining, astrophysics, image and sign
 al processing, hydrodynamics, and more.\n\nMATEDOR is a high-performance n
 umerical library for batched linear algebra subroutines autotuned for mode
 rn processor architectures and system designs. The MATEDOR library include
 s LAPACK-compliant routines that target many small dense problems, tensor,
  and application-specific operations, e.g., for deep-learning. These routi
 nes are constructed as much as possible out of calls to batch BLAS routine
 s and their look-alikes required in sparse computation context.\n\nRegistr
 ation Category: Tech Program Reg Pass, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR