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:20230124T171521Z
LOCATION:C146
DTSTART;TZID=America/Chicago:20221114T113000
DTEND;TZID=America/Chicago:20221114T120000
UID:submissions.supercomputing.org_SC22_sess441_ws_h2rc108@linklings.com
SUMMARY:Accelerating Kernel Ridge Regression with Conjugate Gradient Metho
 d for Large-Scale Data Using FPGA High-Level Synthesis
DESCRIPTION:Workshop\n\nAccelerating Kernel Ridge Regression with Conjugat
 e Gradient Method for Large-Scale Data Using FPGA High-Level Synthesis\n\n
 Alnaser, Langer, Stoll\n\nIn this work, we accelerate the Kernel Ridge Reg
 ression algorithm on an adaptive computing platform to achieve higher perf
 ormance within faster development time by employing a design approach usin
 g high-level synthesis. In order to avoid storing the potentially huge ker
 nel matrix in external memory, the designed accelerator computes the matri
 x on-the-fly in each iteration. Moreover, we overcome the memory bandwidth
  limitation by partitioning the kernel matrix into smaller tiles that are 
 pre-fetched to small local memories and reused multiple times. The design 
 is also parallelized and fully pipelined to accomplish the highest perform
 ance. The final accelerator can be used for any large-scale data without k
 ernel matrix storage limitations and with an arbitrary number of features.
  This work is an important first step towards a library for accelerating d
 ifferent Kernel methods for Machine Learning applications for FPGA platfor
 ms that can be used conveniently from Python with a NumPy interface.\n\nSe
 ssion Format: Recorded\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR