Abstract:To make the problem of multilabel classification with many classes more tractable, in recent years, academia has seen efforts devoted to performing label space dimension reduction (LSDR). Specifically, LSDR encodes high-dimensional label vectors into low-dimensional code vectors lying in a latent space, so as to train predictive models at much lower costs. With respect to the prediction, it performs classification for any unseen instance by recovering a label vector from its predicted code vector via a decoding process. In this paper, we propose a novel method, namely End-to-End Feature-aware label space Encoding (E2FE), to perform LSDR. Instead of requiring an encoding function like most previous works, E2FE directly learns a code matrix formed by code vectors of the training instances in an end-to-end manner. Another distinct property of E2FE is its feature awareness attributable to the fact that the code matrix is learned by jointly maximizing the recoverability of the label space and the predictability of the latent space. Based on the learned code matrix, E2FE further trains predictive models to map instance features into code vectors, and also learns a linear decoding matrix for efficiently recovering the label vector of any unseen instance from its predicted code vector. Theoretical analyses show that both the code matrix and the linear decoding matrix in E2FE can be efficiently learned. Moreover, similar to previous works, E2FE can be specified to learn an encoding function. And it can also be extended with kernel tricks to handle nonlinear correlations between the feature space and the latent space. Comprehensive experiments conducted on diverse benchmark data sets with many classes show consistent performance gains of E2FE over the state-of-the-art methods.

FBLE: A Feature-based Label Enhancement Method for Corporate Bond Credit Rating Prediction

Fast Label Enhancement for Label Distribution Learning

Adversarial Semi-supervised Learning for Corporate Credit Ratings

Dual Contrastive Label Enhancement

Contrastive Label Enhancement

Predicting Label Distribution From Tie-Allowed Multi-Label Ranking

Label Enhancement for Label Distribution Learning

Label Enhancement by Maintaining Positive and Negative Label Relation

Label Distribution Learning Through Exploring Nonnegative Components

End-to-End Feature-Aware Label Space Encoding for Multilabel Classification with Many Classes.

Label Distribution Learning with Label-Specific Features.

Interpretable and Arbitrage-Free Deep Learning for Corporate Bond Pricing

Generalized Label Enhancement with Sample Correlations

Ranking-preserved generative label enhancement

Label Correlations-Based Multi-Label Feature Selection with Label Enhancement

LEFMIFS: Label enhancement and fuzzy mutual information for robust multilabel feature selection

Label distribution learning by utilizing common and label-specific feature fusion space

Unified Framework for Learning with Label Distribution

Generative Label Enhancement with Gaussian Mixture and Partial Ranking.

A multi-level classification based ensemble and feature extractor for credit risk assessment

Corporate bond coupon prediction based on deep learning