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:20250316T231112Z
LOCATION:B306
DTSTART;TZID=America/New_York:20241120T103000
DTEND;TZID=America/New_York:20241120T104500
UID:submissions.supercomputing.org_SC24_sess676_post275@linklings.com
SUMMARY:A Sparse Approach for Translation-Based Training of Knowledge Grap
 h Embeddings
DESCRIPTION:Posters\n\nMd Saidul Hoque Anik and Ariful Azad (Indiana Unive
 rsity)\n\nKnowledge graph (KG) learning offers a powerful framework for ge
 nerating new knowledge and making inferences. Training KG embedding can ta
 ke a significantly long time, especially for larger datasets. Our analysis
  shows that the gradient computation of embedding and vector normalization
  are the dominant functions in the KG embedding training loop. We address 
 this issue by replacing the core embedding computation with SpMM (Sparse-D
 ense Matrix Multiplication) kernels. This allows us to unify multiple scat
 ter (and gather) operations as a single operation, reducing training time 
 and memory usage. Applying this sparse approach in training the TransE mod
 el results in up to 5.7x speedup on the CPU and up to 1.7x speedup on the 
 GPU. Distributing this algorithm on 64 GPUs, we observe up to 3.9x overall
  speedup in each epoch. Our proposed sparse approach can also be extended 
 to accelerate other translation-based models such as TransR and TransH.\n\
 nRegistration Category: Tech Program Reg Pass\n\nSession Chair: Sally Elli
 ngson (University of Kentucky)
END:VEVENT
END:VCALENDAR