Embedding and Clustering Multi-Entity Sequences

authored by: Connor Heaton, Prasenjit Mitra
Abstract: Core to much of modern deep learning is the notion of representation learning, learning representations of things that are useful for performing some task(s) related to those things. Encoder-only language models, for example, learn representations of language useful for performing language-related tasks, often classification. While fruitful in many applications, inherent is the assumption that only one classification is to be made for a particular input. This poses challenges when multiple classifications are to be made about different portions of a single record, such as emotion recognition in conversation (ERC) where the objective is to classify the emotion in each utterance of a dialog. Existing methods for this task typically either involve redundant computation, non-trivial post-processing outside of the core language model backbone, or both. To address this, we generalize recent work for deriving player-specific embeddings from multi-player sequences of events in sport for domain-agnostic application while also enabling it to leverage inter-entity relationships. Seeing the efficacy of the method in regression and classification tasks, we explore how it can be used to cluster player representations, proposing a novel approach for distribution-aware deep-clustering in the absence of labels. We demonstrate how the proposed methods yield state-of-the-art performance on the disparate tasks of ERC in Natural Language Processing (NLP), long-tail partial-label-learning (LT-PLL) in Computer Vision (CV), and player form clustering in sports analytics.
Organisation(s): L3S Research Centre
External Organisation(s): Pennsylvania State University
Type: Article
Journal: IEEE ACCESS
Volume: 12
Pages: 57492-57503
No. of pages: 12
ISSN: 2169-3536
Publication date: 22.04.2024
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: General Computer Science, General Materials Science, General Engineering
Electronic version(s): https://doi.org/10.1109/ACCESS.2024.3391820 (Access: Open)

BibTeX

@article{8a0bbd940e5b4864baa641b761f0248a,
title = "Embedding and Clustering Multi-Entity Sequences",
abstract = "Core to much of modern deep learning is the notion of representation learning, learning representations of things that are useful for performing some task(s) related to those things. Encoder-only language models, for example, learn representations of language useful for performing language-related tasks, often classification. While fruitful in many applications, inherent is the assumption that only one classification is to be made for a particular input. This poses challenges when multiple classifications are to be made about different portions of a single record, such as emotion recognition in conversation (ERC) where the objective is to classify the emotion in each utterance of a dialog. Existing methods for this task typically either involve redundant computation, non-trivial post-processing outside of the core language model backbone, or both. To address this, we generalize recent work for deriving player-specific embeddings from multi-player sequences of events in sport for domain-agnostic application while also enabling it to leverage inter-entity relationships. Seeing the efficacy of the method in regression and classification tasks, we explore how it can be used to cluster player representations, proposing a novel approach for distribution-aware deep-clustering in the absence of labels. We demonstrate how the proposed methods yield state-of-the-art performance on the disparate tasks of ERC in Natural Language Processing (NLP), long-tail partial-label-learning (LT-PLL) in Computer Vision (CV), and player form clustering in sports analytics.",
keywords = "emotion recognition in conversation, long-tail partial-label-learning, Representation learning, sports analytics",
author = "Connor Heaton and Prasenjit Mitra",
note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",
year = "2024",
month = apr,
day = "22",
doi = "10.1109/ACCESS.2024.3391820",
language = "English",
volume = "12",
pages = "57492--57503",
journal = "IEEE ACCESS",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
}