Valve veut utiliser l'IA pour attraper les tricheurs de Counter Strike
It looks like Valve is working on an Artificial Intelligence system to detect and ban cheaters in Counter Strike: Global Offensive.
A Redditor with a moderator-verified Valve_Anti-Cheat account replied to people complaining about the proliferation of âspinbottingâ in Counter Strike: Global Offensive, revealing that Valve is working on an AI-based solution. Spinbotting is a hack that makes it harder for others to hit its user by having his avatar spin rapidly so that its head never stays in one location. While Valve can easily detect this behavior, hackers can just as easily alter the implementation in an eternal cat-and-mouse game.
âSo some bad news: any hard-coded detection of spin-botting leads to an arms race with cheat developers â if they can find the edges of the heuristic youâre using to detect the cheat, the problem comes back. Instead, youâd want to take a machine-learning approach, training (and continuously retraining) a classifier that can detect the differences between cheaters and normal/highly-skilled players.
The process of parsing, training, and classifying player data places serious demands on hardware, which means you want a machine other than the server doing the work. And because you donât know ahead of time who might be using this kind of cheat, youâd have to monitor matches as they take place, from all ten playersâ perspectives.
There are over a million CS:GO matches played every day, so to avoid falling behind youâd need a system capable of parsing and processing every demo of every match from every playerâs perspective, which currently means youâd need a datacenter capable of powering thousands of cpu cores.
The good news is that weâve started this work. An early version of the system has already been deployed and is submitting cases to Overwatch. Since the results have been promising, weâre going to continue this work and expand the system over time.â
Of course the âOverwatchâ mentioned above is Valveâs current system for reporting cheats to be reviewed by human moderators â and not Blizzardâs renowned shooter. Reasonably enough, Valve will use the AI-based system only to flag abuses and leave the final decision to humans â at least, until the AI gets enough training to make its own decisions.
