BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20141120T200000Z
DTEND:20141120T203000Z
LOCATION:292
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: A rapidly emerging theme in large-scale data mining is the use of models based on network theory (also known as graph theory), which allow abstract representation of associations (or “connections”) between entities (or “nodes”). There is growing consensus that basic norms, in spite of their simplicity often represent an acceptable approximation to more complex formulae, subject to appropriate data transformation. L1-norm is well-suited to represent correlation between time-series,  and the Lsup-norm is relevant to situations where emphasis is given to mismatch even in a single parameter. An architecture,  implemented in silicon and available from CogniMem Inc., represents an ideal platform for solving these problems, as it allows massively-parallel calculation of norms between a stored set of reference vectors and vectors that broadcast simultaneously to all processing units. Small clusters of paralleled devices can easily exceed the performance of current multi-GHz CPUs, at a fraction of the power.
SUMMARY:Using Basic Norms for Approximation of Node Associations
PRIORITY:3
END:VEVENT
END:VCALENDAR