SF-SGL: Solver-Free Spectral Graph Learning From Linear Measurements

Learning spectral graph segmentation

We present a general graph learning algorithm for spectral graph partitioning, that allows direct supervised learning of graph structures using hand labeled training examples. The learning algorithm is based on gradient descent in the space of all feasible graph weights. Computation of the gradient involves finding the derivatives of eigenvectors with respect to the graph weight matrix. We show...

Spectral learning of graph distributions

This work draws on previous works regarding spectral learning algorithm for structured data (see [5], [9], [2], [1], [8]). We present an extension of the Hidden Markov Models, called Graphical Weighted Models (GWM), whose purpose is to model distributions over labeled graphs. We describe the spectral algorithm for GWM, which generalizes the previous spectral algorithms for sequences and trees. ...

Rapid estimation of sunflower leaf area from linear measurements

Lean Algebraic Multigrid (LAMG): Fast Graph Laplacian Linear Solver

Laplacian matrices of graphs arise in large-scale computational applications such as semi-supervised machine learning; spectral clustering of images, genetic data and web pages; transportation network flows; electrical resistor circuits; and elliptic partial differential equations discretized on unstructured grids with finite elements. A Lean Algebraic Multigrid (LAMG) solver of the symmetric l...

Analyzing Graph Structure via Linear Measurements

Graphs are an integral data structure for many parts of computation. They are highly effective at modeling many varied and flexible domains, and are excellent for representing the way humans themselves conceive of the world. Nowadays, there is lots of interest in working with large graphs, including social network graphs, “knowledge” graphs, and large bipartite graphs (for example, the Netflix ...