Multiple Instance Regression

This paper introduces multiple instance regression, a variant of multiple regression in which each data point may be described by more than one vector of values for the independent variables. The goals of this work are to (1) understand the computational complexity of the multiple instance regression task and (2) develop an efficient algorithm that is superior to ordinary multiple regression when applied to multiple instance data sets.

• Type: Conference
• Year: 2001
• Where: ICML
• Authors: Soumya Ray, David Page
• Keywords: ICML 2001, Instance Regression Task, Machine Learning, Multiple Instance Regression, Ordinary Multiple Regression