[Ml-stat-talks] Princeton Optimization Seminar: Kush Varshney, Today, 4:30 PM
Amir Ali Ahmadi
a_a_a at princeton.edu
Thu Jan 22 12:09:57 EST 2015
----- Princeton Optimization Seminar -----
DATE: Today, January 22, 2014
LOCATION: Sherrerd Hall room 101
SPEAKER: Kush Varshney, IBM Research
TITLE: Learning Classification Rules via Boolean Compressed Sensing with Applications to Workforce Analytics
ABSTRACT: Motivated by business analytics applications such as identifying employees at risk of voluntary attrition, we propose an interpretable rule-based classification system based on ideas from Boolean compressed sensing. We represent the problem of learning individual conjunctive clauses or individual disjunctive clauses as a Boolean group testing problem, and apply a novel linear programming relaxation to find solutions. We derive results for exact rule recovery which parallel the conditions for exact recovery of sparse signals in the compressed sensing literature. This is an exciting development in rule learning where most prior work focused on heuristic solutions. Furthermore we develop screening methods that use duality theory to dramatically reduce the size of the optimization problem through easily computable certificates that many of the variables must be zero in the optimal solution. We show competitive classification accuracy using the proposed approach.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ml-stat-talks