Local Perturb-and-MAP for Structured Prediction

TL;DR

This paper introduces locPMAP, a local optimization approach for structured prediction that improves inference in CRFs by leveraging a connection to pseudolikelihood, demonstrated across vision tasks.

Contribution

The paper proposes locPMAP, a novel local perturb-and-MAP framework that replaces global optimization with local optimization, enhancing performance and integration in neural networks.

Findings

Consistently improved performance over other approximate inference methods.

Effective integration into fully convolutional networks.

Provides a new perspective on pseudolikelihood in structured prediction.

Abstract

Conditional random fields (CRFs) provide a powerful tool for structured prediction, but cast significant challenges in both the learning and inference steps. Approximation techniques are widely used in both steps, which should be considered jointly to guarantee good performance (a.k.a. "inferning"). Perturb-and-MAP models provide a promising alternative to CRFs, but require global combinatorial optimization and hence they are usable only on specific models. In this work, we present a new Local Perturb-and-MAP (locPMAP) framework that replaces the global optimization with a local optimization by exploiting our observed connection between locPMAP and the pseudolikelihood of the original CRF model. We test our approach on three different vision tasks and show that our method achieves consistently improved performance over other approximate inference techniques optimized to a pseudolikelihood objective. Additionally, we demonstrate that we can integrate our method in the fully convolutional network framework to increase our model's complexity. Finally, our observed connection between locPMAP and the pseudolikelihood leads to a novel perspective for understanding and using pseudolikelihood.