Joint learning of hyperbolic label embeddings for hierarchical multi-label classification

Chatterjee, S. and Maheshwari, A. and Ramakrishnan, G. and J, Saketha Nath (2021) Joint learning of hyperbolic label embeddings for hierarchical multi-label classification. In: 16th Conference of the European Chapter of the Associationfor Computational Linguistics, EACL 2021, 19 April 2021 through 23 April 2021, Virtual, Online.

[img] Text
EACL_2021.pdf - Published Version
Restricted to Registered users only

Download (570kB) | Request a copy


We consider the problem of multi-label classification, where the labels lie in a hierarchy. However, unlike most existing works in hierarchical multi-label classification, we do not assume that the label-hierarchy is known. Encouraged by the recent success of hyperbolic embeddings in capturing hierarchical relations, we propose to jointly learn the classifier parameters as well as the label embeddings. Such a joint learning is expected to provide a twofold advantage: i) the classifier generalises better as it leverages the prior knowledge of existence of a hierarchy over the labels, and ii) in addition to the label co-occurrence information, the label-embedding may benefit from the manifold structure of the input datapoints, leading to embeddings that are more faithful to the label hierarchy. We propose a novel formulation for the joint learning and empirically evaluate its efficacy. The results show that the joint learning improves over the baseline that employs label co-occurrence based pre-trained hyperbolic embeddings. Moreover, the proposed classifiers achieve state-of-the-art generalization on standard benchmarks. We also present evaluation of the hyperbolic embeddings obtained by joint learning and show that they represent the hierarchy more accurately than the other alternatives. The source code of the paper is available here. © 2021 Association for Computational Linguistics

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Conference or Workshop Item (Paper)
Additional Information: We thank Bamdev Mishra and Pratik Jawanpuria (Microsoft India, Hyderabad) for valuable discussions that gave us impetus to work towards this problem. We thank anonymous reviewers for providing constructive feedback. Ayush Maheshwari is supported by a Fellowship from Ekal Foundation ( We are also grateful to IBM Research, India (specifically the IBM AI Horizon Networks - IIT Bombay initiative) for their support and sponsorship.
Uncontrolled Keywords: Co-occurrence informations; Hierarchical multi-label classifications; Hierarchical relations; Joint learning; Manifold structures; Multi label classification; Prior knowledge; State of the art
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: . LibTrainee 2021
Date Deposited: 01 Sep 2022 06:56
Last Modified: 01 Sep 2022 06:56
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 10368 Statistics for this ePrint Item