pith. sign in

arxiv: 1412.4864 · v1 · pith:6WNYMORQnew · submitted 2014-12-16 · 📊 stat.ML · cs.LG· cs.NE

Learning with Pseudo-Ensembles

classification 📊 stat.ML cs.LGcs.NE
keywords pseudo-ensembledropoutnetworkregularizerchildneuralnoiseparent
0
0 comments X
read the original abstract

We formalize the notion of a pseudo-ensemble, a (possibly infinite) collection of child models spawned from a parent model by perturbing it according to some noise process. E.g., dropout (Hinton et. al, 2012) in a deep neural network trains a pseudo-ensemble of child subnetworks generated by randomly masking nodes in the parent network. We present a novel regularizer based on making the behavior of a pseudo-ensemble robust with respect to the noise process generating it. In the fully-supervised setting, our regularizer matches the performance of dropout. But, unlike dropout, our regularizer naturally extends to the semi-supervised setting, where it produces state-of-the-art results. We provide a case study in which we transform the Recursive Neural Tensor Network of (Socher et. al, 2013) into a pseudo-ensemble, which significantly improves its performance on a real-world sentiment analysis benchmark.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.