English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Understanding agent behaviour in Multi-Agent Systems (MAS) is an importantproblem in domains such as autonomous driving, disaster response, and sportsanalytics. Existing MAS problems typically use uniform timesteps withobservations for all agents. In this work, we analyse the problem of agentlocation imputation, specifically posed in environments with non-uniformtimesteps and limited agent observability (~95% missing values). Our approachuses Long Short-Term Memory and Graph Neural Network components to learntemporal and inter-agent patterns to predict the location of all agents atevery timestep. We apply this to the domain of football (soccer) by imputingthe location of all players in a game from sparse event data (e.g., shots andpasses). Our model estimates player locations to within ~6.9m; a ~62% reductionin error from the best performing baseline. This approach facilitatesdownstream analysis tasks such as player physical metrics, player coverage, andteam pitch control. Existing solutions to these tasks often require opticaltracking data, which is expensive to obtain and only available to elite clubs.By imputing player locations from easy to obtain event data, we increase theaccessibility of downstream tasks.

Inferring Player Location in Sports Matches: Multi-Agent Spatial Imputation from Limited Observations