Reducing Label Complexity by Learning From Bags
Sivan Sabato, Nathan Srebreo and Naftali Tishby
In: AISTATS 2010, May 13-15, 2010, Sardinia, Italy.
We consider a supervised learning setting in which the main cost of learning is the number of training labels and one can obtain a single label for a bag of examples, indicating only if a
positive example exists in the bag, as in Multi-Instance Learning. We thus propose to create a training sample of bags, and to use the obtained labels to learn to classify individual examples. We provide a theoretical analysis showing how
to select the bag size as a function of the problem parameters, and prove that if the original labels are distributed unevenly, the number of required labels drops considerably when learning
from bags. We demonstrate that finding a low error separating hyperplane from bags is feasible in this setting using a simple iterative procedure similar to latent SVM. Experiments on synthetic and real data sets demonstrate the success of the