Lower Bounds on Learning Random Structures with Statistical Queries

We show that random DNF formulas, random log-depth decision trees and random deterministic finite acceptors cannot be weakly learned with a polynomial number of statistical queries with respect to an arbitrary distribution on examples.

• Type: Conference
• Year: 2010
• Where: ALT
• Authors: Dana Angluin, David Eisenstat, Leonid Kontorovich, Lev Reyzin
• Keywords: ALT 2010, Log-depth Decision Trees, Machine Learning, Random Deterministic Finite, Random Dnf Formulas