Multi-Agent Deep-Learning Based Comparative Analysis of Team Sport Trajectories

Abstract

Computational analysis of multi-agent trajectories is a fundamental issue in the study of real-world biological agents. For trajectory analysis, combining movement data with labels (e.g., whether a team scores in a ball game) can provide additional insights compared to relying only on trajectory data. However, existing deep-learning-based methods consider only single-agent animal trajectories, and cannot be directly applied to multi-agent trajectories in sports. In this paper, a comparative analysis method to analyze multi-agent trajectories in ball games is proposed. A neural network approach using multi-agent motion characteristics (e.g., the distances between agents and objects) as the input is adopted, which is based on an attention mechanism designed to automatically detect segments in trajectories that are characteristic of a given class. This enables us to understand differences between classes by highlighting segmented trajectories and which variables correlate with the given labels. The effectiveness of our approach was verified by comparing various baselines with effective/ineffective attack labels and goal/non-goal labels using different sizes of the dataset. The effectiveness of our method is also demonstrated through a use case that analyzes the attacking plays in an NBA dataset.

Teaming behavior in adversarial scenarios

“…One difficulty is that these games are far from the simplistic mathematical games, natively featuring a complex panoply of phenomena, including heterogeneous agents that have differing psychologies and who take on distinct roles within the team strategy. As such, some early work is in the direction of using machine learning techniques to learn player-specific action templates, which can be leveraged to understand common play concepts (Miller and Bornn, 2017;Ziyi et al, 2023). Other works aim to give a glimpse into domain knowledge of high-level intuitive concepts which may be useful to guide our development of multiagent robot team strategies (Fernandez and Bornn, 2018).…”

The Learning Trajectory Based on STEM of Elementary School Pupils’ in Solving Proportion Material: Didactical Design-Research

“…The concept of ratios is one of the fundamental mathematical concepts that are important and commonly used in daily life. With the use of STEM-based learning media, students can easily understand and apply the concept of ratios in real-life situations (Ziyi, 2023). Overall, the use of STEM-based learning media in mathematics learning can help students acquire a better understanding of mathematical concepts and skills, as well as improve their problem-solving abilities effectively.…”