most citedAn initial alignment between neural network and target is needed for gradient descent to learn

2 citations · 4 across the 2 of their papers we have counts for

collaborators

2 papers

stat.ML2022★ 2 cited

Using Mixed-Effects Models to Learn Bayesian Networks from Related Data Sets

Marco Scutari, Christopher Marquis, Laura Azzimonti

We commonly assume that data are a homogeneous set of observations when learning the structure of Bayesian networks. However, they often comprise different data sets that are relat…

cs.LG2022★ 2 cited

An initial alignment between neural network and target is needed for gradient descent to learn

Emmanuel Abbe, Elisabetta Cornacchia, Jan Hązła +1

This paper introduces the notion of ``Initial Alignment'' (INAL) between a neural network at initialization and a target function. It is proved that if a network and a Boolean targ…