Propensity-scored Probabilistic Label Trees

Abstract

Extreme multi-label classification (XMLC) refers to the task of tagging instances with small subsets of relevant labels coming from an extremely large set of all possible labels. Recently, XMLC has been widely applied to diverse web applications such as automatic content labeling, online advertising, or recommendation systems. In such environments, label distribution is often highly imbalanced, consisting mostly of very rare tail labels, and relevant labels can be missing. As a remedy to these problems, the propensity model has been introduced and applied within several XMLC algorithms. In this work, we focus on the problem of optimal predictions under this model for probabilistic label trees, a popular approach for XMLC problems. We introduce an inference procedure, based on the $A^{\ast }$-search algorithm, that efficiently finds the optimal solution, assuming that all probabilities and propensities are known. We demonstrate the attractiveness of this approach in a wide empirical study on popular XMLC benchmark datasets.

The extended version of SIGIR '21 Short Research Paper

References in corpus (7)

• Bonsai -- Diverse and Shallow Trees for Extreme Multi-label Classification
• AttentionXML: Label Tree-based Attention-Aware Deep Model for High-Performance Extreme Multi-Label Text Classification
• Conditional Probability Tree Estimation Analysis and Algorithms
• Surrogate Regret Bounds for Bipartite Ranking via Strongly Proper Losses
• A no-regret generalization of hierarchical softmax to extreme multi-label classification
• PECOS: Prediction for Enormous and Correlated Output Spaces
• Probabilistic Label Trees for Extreme Multi-label Classification

Cited by in corpus (1)

• On Missing Labels, Long-tails and Propensities in Extreme Multi-label Classification