Robocalypse Now: Using Deep Learning to Combat Cheating in Counter-Strike: Global Offensive

Channel:GDC Festival of Gaming · Length:1:05:25 · Published:2020-06-18 · Views:35470

Watch on YouTube

Summary

Valve's John McDonald presents at GDC 2018 on how Valve uses deep learning (VACnet) to detect cheaters in Counter-Strike: Global Offensive. He covers training on human-reviewed replays and how model output is confirmed by manual review.

Why HN recommends it

The go-to link in any anti-cheat thread for a concrete, well-delivered account of machine-learning cheat detection in production.

The presenter delivered the problem and solution approach so well I'd still like to work on something like this some time. — HN, 2025

Machine learning & deep learning · Game programming & engines · Gaming culture & analysis

Subtitles (excerpt)

Uh hi. So, we have a little bit of business before we get started. I'm supposed to ask you to pretty please mute your cell phones so they don't go off. And also to make sure to fill out your surveys when you're done or when we're done. They're super helpful for GDC to figure out who they should invite back and they're also super useful for me to figure out if you like the way I presented and you like what I presented. So, please take a moment when we're done to do that. All right. That's That's the stuff out of the way. Let's get started. So, my name is John McDonald. I'm here from Valve and today I wanted to come and talk a little bit about how we've been using deep learning to effectively combat cheating in CS:GO. Spoiler alert, it works pretty well. We're pretty happy with it. So, that's a bunch of stuff. We have a whole hour, so we're going to We're going to cover a lot of ground. I'm going to spend a little bit of time, probably fair warning, probably about 15 minutes kind of talking about the state of CS:GO with respect to cheating, other things that we do in this space. And then we'll dive into sort of the pitch that I was making to folks around Valve in 2016 for why we should maybe give this try.

On Hacker News

Woah, cool to see a bit of how it all works. Seems similar to basic antivirus signature checking? As a side note, a few years back, they had a GDC talk about a (then-new) machine learning based approach to catching cheaters. Regardless of how it worked out, I found it fascinating. The presenter delivered the problem and solution approach so well I'd still like to work on something like this some … — schmorptron, 2025-06
Unfortunately the Overwatch system has been disabled for years, and we currently have no reason to think it will be reopened for CS2. I take this opportunity to share this great talk about Valve's usage of deep learning to fight cheating in CS:GO: [VIDEO] — magic123_, 2024-08
Valve has been at this for a long time... They started with a system that lets players review games and try to understand if a player is a cheat or not. Games are anonymized, and you get to rewatch the replay to try to make a determination They can then use that data to train their NN to help detect which players / games should be fed into this manual review process I assume now the product is pr… — arwineap, 2022-11
Or server-side automated review using machine learning models, which is what Valve is trying with "vacnet" (with then manual review to confirm the model's results are accurate - something most tech companies should learn from) Great talk on it here: [VIDEO] — kllrnohj, 2022-08
Valve do this, here's a good talk: Robocalypse Now: Using Deep Learning to Combat Cheating in Counter-Strike: Global Offensive [VIDEO] — crtasm, 2022-02

Related videos