MLMay 20, 2016

Learning to Discover Sparse Graphical Models

arXiv:1605.06359v334 citations
Originality Incremental advance
AI Analysis

This addresses the challenge of structure recovery in graphical models for domains like genetics and brain imaging, offering a more adaptable and efficient approach, though it is incremental as it builds on existing methods by using synthetic training data.

The paper tackles the problem of discovering sparse graphical models from observational data by proposing a neural network that maps covariance matrices to graph structures, achieving superior performance on genetics, brain imaging, and simulation data compared to analytical methods.

We consider structure discovery of undirected graphical models from observational data. Inferring likely structures from few examples is a complex task often requiring the formulation of priors and sophisticated inference procedures. Popular methods rely on estimating a penalized maximum likelihood of the precision matrix. However, in these approaches structure recovery is an indirect consequence of the data-fit term, the penalty can be difficult to adapt for domain-specific knowledge, and the inference is computationally demanding. By contrast, it may be easier to generate training samples of data that arise from graphs with the desired structure properties. We propose here to leverage this latter source of information as training data to learn a function, parametrized by a neural network that maps empirical covariance matrices to estimated graph structures. Learning this function brings two benefits: it implicitly models the desired structure or sparsity properties to form suitable priors, and it can be tailored to the specific problem of edge structure discovery, rather than maximizing data likelihood. Applying this framework, we find our learnable graph-discovery method trained on synthetic data generalizes well: identifying relevant edges in both synthetic and real data, completely unknown at training time. We find that on genetics, brain imaging, and simulation data we obtain performance generally superior to analytical methods.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes