Learning to Move Like Professional Counter-Strike Players

Simple item page
dc.contributor.authorDurst, Daviden_US
dc.contributor.authorXie, Fengen_US
dc.contributor.authorSarukkai, Vishnuen_US
dc.contributor.authorShacklett, Brennanen_US
dc.contributor.authorFrosio, Iurien_US
dc.contributor.authorTessler, Chenen_US
dc.contributor.authorKim, Joohwanen_US
dc.contributor.authorTaylor, Carlyen_US
dc.contributor.authorBernstein, Gilberten_US
dc.contributor.authorChoudhury, Sanjibanen_US
dc.contributor.authorHanrahan, Paten_US
dc.contributor.authorFatahalian, Kayvonen_US
dc.contributor.editorSkouras, Melinaen_US
dc.contributor.editorWang, Heen_US
dc.date.accessioned2024-08-20T08:42:40Z
dc.date.available2024-08-20T08:42:40Z
dc.date.issued2024
dc.description.abstractIn multiplayer, first-person shooter games like Counter-Strike: Global Offensive (CS:GO), coordinated movement is a critical component of high-level strategic play. However, the complexity of team coordination and the variety of conditions present in popular game maps make it impractical to author hand-crafted movement policies for every scenario. We show that it is possible to take a data-driven approach to creating human-like movement controllers for CS:GO. We curate a team movement dataset comprising 123 hours of professional game play traces, and use this dataset to train a transformer-based movement model that generates human-like team movement for all players in a ''Retakes'' round of the game. Importantly, the movement prediction model is efficient. Performing inference for all players takes less than 0.5 ms per game step (amortized cost) on a single CPU core, making it plausible for use in commercial games today. Human evaluators assess that our model behaves more like humans than both commercially-available bots and procedural movement controllers scripted by experts (16% to 59% higher by TrueSkill rating of ''human-like''). Using experiments involving in-game bot vs. bot self-play, we demonstrate that our model performs simple forms of teamwork, makes fewer common movement mistakes, and yields movement distributions, player lifetimes, and kill locations similar to those observed in professional CS:GO match play.en_US
dc.description.number8
dc.description.sectionheadersCharacter Animation II: Control
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume43
dc.identifier.doi10.1111/cgf.15173
dc.identifier.issn1467-8659
dc.identifier.pages12 pages
dc.identifier.urihttps://doi.org/10.1111/cgf.15173
dc.identifier.urihttps://diglib.eg.org/handle/10.1111/cgf15173
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectCCS Concepts: Software and its engineering → Interactive games; Computing methodologies → Learning from demonstrations
dc.subjectSoftware and its engineering → Interactive games
dc.subjectComputing methodologies → Learning from demonstrations
dc.titleLearning to Move Like Professional Counter-Strike Playersen_US
Files
Original bundle
Now showing 1 - 3 of 3
No Thumbnail Available
Name:
v43i8_cgf15173.pdf
Size:
12.08 MB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
learningtomovelikeprofessionalcounter-strikeplayers-supplemental.pdf
Size:
322.06 KB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
user_study.zip
Size:
245.53 MB
Format:
Zip file
Download
Collections
43-Issue 8