In a recent post on Reddit, Valve appears to be bringing in AI (Artificial Intelligence) to help combat the not uncommon cheating in Counter-Strike: Global Offensive.
No doubt if you have ever played Counter-Strike: Global Offensive, you have come across various cheaters, glitchers, and hackers who have a tendency to make the game less fun for everyone except themselves. While Valve does have a fairly good anti-cheat system (Valve Anti Cheat/VAC), code monkeys still find ways around the various tools Valve has to combat hacking.
One of the recent questions brought up on Reddit was why Valve does not automatically ban players who use a 360 spin’n’shoot hack (spinbot) since it should be fairly easy to detect? The answer provided by a Valve representative was more telling than expected regarding Counter-Strike: Global Offensive cheat detecting mechanics.
“So some bad news: any hard-coded detection of spin-botting leads to an arms race with cheat developers – if they can find the edges of the heuristic you’re using to detect the cheat, the problem comes back. Instead, you’d 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.”
Anyone who plays an online FPS long enough knows that while a hack may die down for a while, someone, somewhere is going to find a way around it. The machine-learning approach is a reference to a form of computer AI where the AI can actual learn and develop a “sense” of what hacking constitutes based on the input parameters from the game developers. And the Valve official had more to say on the requirements such a system would need.
“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 don’t know ahead of time who might be using this kind of cheat, you’d have to monitor matches as they take place, from all ten players’ perspectives.
“There are over a million CS:GO matches played every day, so to avoid falling behind you’d need a system capable of parsing and processing every demo of every match from every player’s perspective, which currently means you’d need a data center capable of powering thousands of CPU cores.
“The good news is that we’ve 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, we’re going to continue this work and expand the system over time.”
Part of this statement is not necessarily accurate, specifically the idea that the system would need to monitor every Counter-Strike match from every player’s perspective. To wit, the purpose of the machine-learning/AI concept would be for the system to “learn” what is hacking and what is not. Since millions matches occur every day, and every match has a significant number of viewpoints, the system would not need to watch every match to begin learning what a hack constitutes.
Furthermore, while gamers would no doubt prefer the system (henceforth, Valvenet) to monitor every Counter-Strike match and every player for potential cheaters, a properly designed system wouldn’t need to do so. The data streaming in from the entirety of the match would indicate what actions are being taken throughout, and the system would develop a series of flags which warrant further investigation.
Valvenet also would not need to monitor each match, but rather take a large sampling of each day’s batch to analyze. The cheat report tool would also add another trigger for Valvenet to investigate, limiting the necessary computational and processing power such a system would require.
Fortunately, Valve seems to have a plan in place to help counteract cheating in Counter-Strike: Global Offensive.
So what are your thoughts on Valve’s plan to bring machine learning/AI to bear against cheating/hacking in Counter-Strike: Global Offensive? Tell us what you think in the comments section below!
[Featured Image by Counter-Strike: Global Offensive/Steam]