Constraint-aware Learning of Probabilistic Sequential Models for Multi-Label Classification

TL;DR

This paper presents a constraint-aware probabilistic sequential modeling approach for multi-label classification, effectively exploiting and enforcing logical constraints among labels to improve joint distribution modeling.

Contribution

It introduces an architecture combining individual label classifiers with a sequential model that captures label correlations and constraints, enhancing multi-label classification.

Findings

The model can exploit constraints during training.

It effectively enforces constraints during inference.

Demonstrates improved modeling of label correlations.

Abstract

We investigate multi-label classification involving large sets of labels, where the output labels may be known to satisfy some logical constraints. We look at an architecture in which classifiers for individual labels are fed into an expressive sequential model, which produces a joint distribution. One of the potential advantages for such an expressive model is its ability to modelling correlations, as can arise from constraints. We empirically demonstrate the ability of the architecture both to exploit constraints in training and to enforce constraints at inference time.