A Foundation Model for Soccer

Ethan Baron, Daniel Hocevar, Zach Salehe
7/18/2024
cs.LG

Official code: danielhocevar/Foundation-Model-for-Soccer (20 stars, last commit 2024-04-19).

Abstract

We propose a foundation model for soccer, which is able to predict subsequent actions in a soccer match from a given input sequence of actions. As a proof of concept, we train a transformer architecture on three seasons of data from a professional soccer league. We quantitatively and qualitatively compare the performance of this transformer architecture to two baseline models: a Markov model and a multi-layer perceptron. Additionally, we discuss potential applications of our model. We provide an open-source implementation of our methods at https://github.com/danielhocevar/Foundation-Model-for-Soccer.

View on arXivView PDF

Code Implementations(1)

danielhocevar/Foundation-Model-for-SoccerOfficial100%
201Python, Jupyter NotebookFeb 23, 20242 years ago

Similar Papers

Predicting soccer matches with complex networks and machine learning

84% match2024

FootBots: A Transformer-based Architecture for Motion Prediction in Soccer

82% match2024

TranSPORTmer: A Holistic Approach to Trajectory Understanding in Multi-Agent Sports

81% match2024

Forecasting Events in Soccer Matches Through Language

80% match1 repo192024

SoccerDiffusion: Toward Learning End-to-End Humanoid Robot Soccer from Gameplay Recordings

79% match2025

A Graph Neural Network deep-dive into successful counterattacks

79% match2024

Cite this paper

@article{baron2024foundation,
  title  = {A Foundation Model for Soccer},
  author = {Ethan Baron and Daniel Hocevar and Zach Salehe},
  year   = {2024},
  eprint = {2407.14558},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.14558v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.14558.svg)](https://codeofpaper.com/paper/2407.14558)

Discussion