It can detect if a player is using cheats based on their behaviour and the actions they perform on the map.
According to GosuAI, there is at least one cheater in 12% of our games.
The GosuAI algorithm is based on machine learning and currently isn’t able to detect all kind of cheats. They analyzed approximately 200k matches played by about 1 million unique players, and learned to spot cheaters who use zoomout, autostop of creeps, and items auto drop.
To detect cheating we extracted mouse movements and all the actions player perform on a map from replays. To analyse player behavior we aggregate several statistics like the distance between the mouse position on a screen and the place where the player’s real action was registered. Then we use machine learning technique called anomaly detection: the algorithm learns from a sample of “pure” matches (manually checked by our Dota 2 experts for the absence of cheating) and then able to detect “suspicious” deviations from the norm.”
Camera hack. The most popular cheat, used in 12.24% of the matches. With it the player can look quickly at distant places on the map, gaining a vision awareness advantage. Note that the hack won't detect ultra-wide monitor users as cheaters.
Automatic creep blocking. Only 0.1% of the player base uses it. A perfect creep blocking at the start of the match can give you an enormous boost, particularly in the mid lane.
Automatic items dropping. Used by at least 0.035% players. For example, dropping and picking back the Arcane Boots while using the Soul Ring gives you 75 extra mana.
