Discriminative Embeddings of Latent Variable Models for Structured Data

TL;DR

This paper introduces structure2vec, a scalable method for embedding structured data into feature spaces using discriminative learning, outperforming traditional kernel methods in speed, model size, and predictive accuracy.

Contribution

It presents a novel embedding approach that integrates latent variable models with discriminative training, enabling scalable and efficient structured data analysis.

Findings

Runs twice as fast on large datasets

Produces models 10,000 times smaller

Achieves state-of-the-art predictive performance

Abstract

Kernel classifiers and regressors designed for structured data, such as sequences, trees and graphs, have significantly advanced a number of interdisciplinary areas such as computational biology and drug design. Typically, kernels are designed beforehand for a data type which either exploit statistics of the structures or make use of probabilistic generative models, and then a discriminative classifier is learned based on the kernels via convex optimization. However, such an elegant two-stage approach also limited kernel methods from scaling up to millions of data points, and exploiting discriminative information to learn feature representations. We propose, structure2vec, an effective and scalable approach for structured data representation based on the idea of embedding latent variable models into feature spaces, and learning such feature spaces using discriminative information. Interestingly, structure2vec extracts features by performing a sequence of function mappings in a way similar to graphical model inference procedures, such as mean field and belief propagation. In applications involving millions of data points, we showed that structure2vec runs 2 times faster, produces models which are $10,000$ times smaller, while at the same time achieving the state-of-the-art predictive performance.