Masked Autoencoder Pretraining for Event Classification in Elite Soccer

Yannick Rudolph; Ulf Brefeld

doi:10.1007/978-3-031-53833-9_3

Masked Autoencoder Pretraining for Event Classification in Elite Soccer

Yannick Rudolph^*
, Ulf Brefeld

^*Corresponding author for this work

Professorship of Information Systems, in particular Machine Learning

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

Abstract

We show that pretraining transformer models improves the performance on supervised classification of tracking data from elite soccer. Specifically, we propose a novel self-supervised masked autoencoder for multiagent trajectories. In contrast to related work, our approach is significantly simpler, has no necessity for handcrafted features and inherently allows for permutation invariance in downstream tasks.

Original language	English
Title of host publication	Machine Learning and Data Mining for Sports Analytics : 10th International Workshop, MLSA 2023, Revised Selected Papers
Editors	Ulf Brefeld, Jesse Davis, Jan Van Haaren, Albrecht Zimmermann
Number of pages	12
Place of Publication	Cham
Publisher	Springer Nature Switzerland AG
Publication date	26.02.2024
Pages	24-35
ISBN (Print)	978-3-031-53832-2
ISBN (Electronic)	978-3-031-53833-9
DOIs	https://doi.org/10.1007/978-3-031-53833-9_3
Publication status	Published - 26.02.2024
Event	10th International Workshop on Machine Learning and Data Mining for Sports Analytics - MLSA 2023 - Turin, Italy Duration: 18.09.2023 → 18.09.2023 Conference number: 10 https://dtai.cs.kuleuven.be/events/MLSA23/ http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=173575&copyownerid=49896

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

Research areas and keywords

Factorized transformer architecture
Masked autoencoder
Multiagent trajectories
Self-supervised learning
Soccer
Tracking data
Informatics
Business informatics

Access to Document

10.1007/978-3-031-53833-9_3

Masked autoencoder for multiagent trajectories
Rudolph, Y. & Brefeld, U., 02.2025, In: Machine Learning. 114, 2, 18 p., 44.
Research output: Journal contributions › Journal articles › Research › peer-review

Open Access

Masked Autoencoder Pretraining for Event Classification in Elite Soccer
Rudolph, Y. (Speaker) & Brefeld, U. (Coauthor)
18.09.2023
Activity: Talk or presentation › talk or presentation in privat or public events › Research

Cite this

Rudolph, Y., & Brefeld, U. (2024). Masked Autoencoder Pretraining for Event Classification in Elite Soccer. In U. Brefeld, J. Davis, J. Van Haaren, & A. Zimmermann (Eds.), Machine Learning and Data Mining for Sports Analytics: 10th International Workshop, MLSA 2023, Revised Selected Papers (pp. 24-35). (Communications in Computer and Information Science; Vol. 2035). Springer Nature Switzerland AG. https://doi.org/10.1007/978-3-031-53833-9_3

Rudolph, Yannick ; Brefeld, Ulf. / Masked Autoencoder Pretraining for Event Classification in Elite Soccer. Machine Learning and Data Mining for Sports Analytics: 10th International Workshop, MLSA 2023, Revised Selected Papers. editor / Ulf Brefeld ; Jesse Davis ; Jan Van Haaren ; Albrecht Zimmermann. Cham : Springer Nature Switzerland AG, 2024. pp. 24-35 (Communications in Computer and Information Science).

@inbook{0ca682cc4633469e8d5e809ba430275d,

title = "Masked Autoencoder Pretraining for Event Classification in Elite Soccer",

abstract = "We show that pretraining transformer models improves the performance on supervised classification of tracking data from elite soccer. Specifically, we propose a novel self-supervised masked autoencoder for multiagent trajectories. In contrast to related work, our approach is significantly simpler, has no necessity for handcrafted features and inherently allows for permutation invariance in downstream tasks.",

keywords = "Factorized transformer architecture, Masked autoencoder, Multiagent trajectories, Self-supervised learning, Soccer, Tracking data, Informatics, Business informatics",

author = "Yannick Rudolph and Ulf Brefeld",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; 10th International Workshop on Machine Learning and Data Mining for Sports Analytics - MLSA 2023, MLSA 2023 ; Conference date: 18-09-2023 Through 18-09-2023",

year = "2024",

month = feb,

day = "26",

doi = "10.1007/978-3-031-53833-9\_3",

language = "English",

isbn = "978-3-031-53832-2",

series = "Communications in Computer and Information Science",

publisher = "Springer Nature Switzerland AG",

pages = "24--35",

editor = "Ulf Brefeld and Jesse Davis and \{Van Haaren\}, Jan and Albrecht Zimmermann",

booktitle = "Machine Learning and Data Mining for Sports Analytics",

address = "Switzerland",

url = "https://dtai.cs.kuleuven.be/events/MLSA23/ , http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=173575\&copyownerid=49896",

}

Rudolph, Y & Brefeld, U 2024, Masked Autoencoder Pretraining for Event Classification in Elite Soccer. in U Brefeld, J Davis, J Van Haaren & A Zimmermann (eds), Machine Learning and Data Mining for Sports Analytics: 10th International Workshop, MLSA 2023, Revised Selected Papers. Communications in Computer and Information Science, vol. 2035, Springer Nature Switzerland AG, Cham, pp. 24-35, 10th International Workshop on Machine Learning and Data Mining for Sports Analytics - MLSA 2023, Turin, Italy, 18.09.23. https://doi.org/10.1007/978-3-031-53833-9_3

Masked Autoencoder Pretraining for Event Classification in Elite Soccer. / Rudolph, Yannick ; Brefeld, Ulf.
Machine Learning and Data Mining for Sports Analytics: 10th International Workshop, MLSA 2023, Revised Selected Papers. ed. / Ulf Brefeld; Jesse Davis; Jan Van Haaren; Albrecht Zimmermann. Cham: Springer Nature Switzerland AG, 2024. p. 24-35 (Communications in Computer and Information Science; Vol. 2035).

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

TY - CHAP

T1 - Masked Autoencoder Pretraining for Event Classification in Elite Soccer

AU - Rudolph, Yannick

AU - Brefeld, Ulf

N1 - Conference code: 10

PY - 2024/2/26

Y1 - 2024/2/26

N2 - We show that pretraining transformer models improves the performance on supervised classification of tracking data from elite soccer. Specifically, we propose a novel self-supervised masked autoencoder for multiagent trajectories. In contrast to related work, our approach is significantly simpler, has no necessity for handcrafted features and inherently allows for permutation invariance in downstream tasks.

AB - We show that pretraining transformer models improves the performance on supervised classification of tracking data from elite soccer. Specifically, we propose a novel self-supervised masked autoencoder for multiagent trajectories. In contrast to related work, our approach is significantly simpler, has no necessity for handcrafted features and inherently allows for permutation invariance in downstream tasks.

KW - Factorized transformer architecture

KW - Masked autoencoder

KW - Multiagent trajectories

KW - Self-supervised learning

KW - Soccer

KW - Tracking data

KW - Informatics

KW - Business informatics

UR - https://www.scopus.com/pages/publications/85187650562

UR - https://www.mendeley.com/catalogue/df720300-8c0a-37b1-a340-cb810b405c19/

U2 - 10.1007/978-3-031-53833-9_3

DO - 10.1007/978-3-031-53833-9_3

M3 - Article in conference proceedings

AN - SCOPUS:85187650562

SN - 978-3-031-53832-2

T3 - Communications in Computer and Information Science

SP - 24

EP - 35

BT - Machine Learning and Data Mining for Sports Analytics

A2 - Brefeld, Ulf

A2 - Davis, Jesse

A2 - Van Haaren, Jan

A2 - Zimmermann, Albrecht

PB - Springer Nature Switzerland AG

CY - Cham

T2 - 10th International Workshop on Machine Learning and Data Mining for Sports Analytics - MLSA 2023

Y2 - 18 September 2023 through 18 September 2023

ER -

Rudolph Y , Brefeld U. Masked Autoencoder Pretraining for Event Classification in Elite Soccer. In Brefeld U, Davis J, Van Haaren J, Zimmermann A, editors, Machine Learning and Data Mining for Sports Analytics: 10th International Workshop, MLSA 2023, Revised Selected Papers. Cham: Springer Nature Switzerland AG. 2024. p. 24-35. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-53833-9_3

Masked Autoencoder Pretraining for Event Classification in Elite Soccer

Abstract

Bibliographical note

Research areas and keywords

Access to Document

Fingerprint

Research output

Masked autoencoder for multiagent trajectories

Activities

Masked Autoencoder Pretraining for Event Classification in Elite Soccer

Cite this