Partial Inference in Structured Prediction

In this work, we examine the partial inference problem in the context of structured prediction. Using a generative model approach, we consider the task of maximizing a score function with unary and pairwise potentials in the space of labels on graphs. Employing a two-stage convex optimization algorithm for label recovery, we analyze the conditions under which a majority of the labels can be recovered. We introduce a novel perspective on the Karush-Kuhn-Tucker (KKT) conditions and primal and dual construction, and provide statistical and topological requirements for partial recovery with provable guarantees. The full-length paper with detailed proofs of our novel theoretical claims can be accessed from https://arxiv.org/abs/2306.03949.