It is shown that \(AC^0\) circuits can be learned by the Low Degree Algorithm in quasi-polynomial-time under the uniform distribution due to Linial, ...

It is well known that \(AC^0\) circuits can be learned by the Low Degree Algorithm in quasi-polynomial-time under the uniform distribution due to Linial, ...

An important and long-standing question in computational learning theory is how to learn. AC0 circuits with respect to any distribution (i.e. PAC learning). All ...

Two extensions of the Linial, Mansour, Nisan 0 learning algorithm are presented. The LMN method works when input examples are drawn uniformly.

In this paper we develop two extensions of the LMN learn- ing algorithm which produce good approximating functions when samples are drawn according to unknown ...

It is well known that A C 0 circuits can be learned by the Low Degree Algorithm in quasi-polynomial-time under the uniform distribution due to Linial, ...

In 2002 Jackson [JKS02] asked whether AC 0 circuits augmented with a threshold gate at the output can be efficiently learned from uniform random examples.

Abstract. In many problems, the measured variables (e.g., image pixels) are just mathematical functions of the hidden causal variables (e.g., the underlying.

Approximating a divergence between two probability distributions from their sam- ples is a fundamental challenge in statistics, information theory, and machine ...

In this model the goal is to approximately infer an un- known target concept that belongs to some known con- cept class using positive.