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:20181221T160904Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post142@linklings.com
SUMMARY:Sol: Transparent Neural Network Acceleration Platform
DESCRIPTION:Poster\nTech Program Reg Pass, Exhibits Reg Pass\n\nSol: Trans
 parent Neural Network Acceleration Platform\n\nWeber\n\nWith the usage of 
 neural networks in a wide range of application fields, the necessity to ex
 ecute these efficiently on high performance hardware is one of the key pro
 blems for artificial intelligence (AI) framework providers. More and more 
 new specialized hardware types and corresponding libraries appear from var
 ious manufacturers. The biggest problem arising is that these libraries us
 ually are only supported by a very limited set of AI frameworks and intero
 perability can become an issue. In this extended abstract we present Sol, 
 a transparent middleware for neural network acceleration. Sol comes with a
 n optimizing compiler engine, allowing to use device specific libraries an
 d to implement own optimizations, that can be leveraged on all target devi
 ces. In contrast to other projects Sol explicitly aims at optimizing predi
 ction and training of neural networks.
URL:https://sc18.supercomputing.org/presentation/?id=post142&sess=sess324
END:VEVENT
END:VCALENDAR