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:20181221T160730Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181113T160000
DTEND;TZID=America/Chicago:20181113T163000
UID:submissions.supercomputing.org_SC18_sess212_pap273@linklings.com
SUMMARY:Distributed Memory Sparse Inverse Covariance Matrix Estimation on 
 High-Performance Computing Architectures
DESCRIPTION:Paper\nAlgorithms, Graph Algorithms, Linear Algebra, Machine L
 earning, Sparse Computation, Tech Program Reg Pass\n\nDistributed Memory S
 parse Inverse Covariance Matrix Estimation on High-Performance Computing A
 rchitectures\n\nEftekhari, Bollhöfer, Schenk\n\nWe consider the problem of
  estimating sparse inverse covariance matrices for high-dimensional datase
 ts using the l1-regularized Gaussian maximum likelihood method. This task 
 is particularly challenging as the required computational resources increa
 se superlinearly with the dimensionality of the dataset. We introduce a pe
 rformant and scalable algorithm which builds on the current advancements o
 f second-order, maximum likelihood methods. The routine leverages the intr
 insic parallelism in the linear algebra operations and exploits the underl
 ying sparsity of the problem. The computational bottlenecks are identified
  and the respective subroutines are parallelized using an MPI-OpenMP appro
 ach. Experiments conducted on a Cray XC50 system at the Swiss National Sup
 ercomputing Center show that, in comparison to the state-of-the-art algori
 thms, the proposed routine provides significant strong scaling speedup wit
 h ideal scalability up to 128 nodes. The developed framework is used to es
 timate the sparse inverse covariance matrix of both synthetic and real-wor
 ld datasets with up to 10 million dimensions.
URL:https://sc18.supercomputing.org/presentation/?id=pap273&sess=sess212
END:VEVENT
END:VCALENDAR