Learning DNF from Random Walks

We consider a model of learning Boolean functions from examples generated by a uniform random walk on {0, 1}n . We give a polynomial time algorithm for learning decision trees and DNF formulas in this model. This is the first efficient algorithm for learning these classes in a natural passive learning model where the learner has no influence over the choice of examples used for learning. ∗ Supported by a Miller Postdoctoral Fellowship. † Supported by NSF grant 99-12342. ‡ Supported by an NSF Mathematical Sciences Postdoctoral Fellowship and by NSF grant CCR-98-77049. Most of this research was performed while at Harvard University. 

• Type: Conference
• Year: 2003
• Where: FOCS
• Authors: Nader H. Bshouty, Elchanan Mossel, Ryan O'Donnell, Rocco A. Servedio
• Keywords: FOCS 2003, NSF Grant, Passive Learning Model, Postdoctoral Fellowship, Theoretical Computer Science