Learning to Rate Player Positioning in Soccer

被引：37

作者：

Dick, Uwe ^{[1
]}

Brefeld, Ulf ^{[1
]}

机构：

[1] Leuphana Univ, Inst Informat Syst, Luneburg, Germany

来源：

BIG DATA | 2019年 / 7卷 / 01期

关键词：

deep learning; spatiotemporal data; reinforcement learning; scoring function;

D O I：

10.1089/big.2018.0054

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

We investigate how to learn functions that rate game situations on a soccer pitch according to their potential to lead to successful attacks. We follow a purely data-driven approach using techniques from deep reinforcement learning to valuate multiplayer positionings based on positional data. Empirically, the predicted scores highly correlate with dangerousness of actual situations and show that rating of player positioning without expert knowledge is possible.

引用

页码：71 / 82

页数：12

共 30 条

[1]

[Anonymous], 2016, PROC INT C MACH LEAR

[2]

[Anonymous], 2014, INT C MACH LEARN

[3] Approximate policy iteration: A survey and some new methods [J].