Current State of Research

[DevinTDHa]

Hi Everyone,

I think the idea of the project is very cool (although it will be very hard to achieve) and would like to start a discussion on resources in the current field.

• I was aware, that Valve is using VACnet which is also a Deep Learning Approach.
  J. McDonald. Robocalypse now: Using deep learning to combat cheating in ’Counter-strike: Global offensive’. Presentation at the Game Developers Conference (GDC), 2018.

• Cheats for minor performance boosts (as some top players are suspected in using) are very hard to detect:
  Witschel, Tim, and Christian Wressnegger. “Aim Low, Shoot High: Evading Aimbot Detectors by Mimicking User Behavior.” Proceedings of the 13th European Workshop on Systems Security, Apr. 2020, pp. 19–24. arXiv.org, https://doi.org/10.1145/3380786.3391397.

• There is a very recent paper on another deep learning based cheat detection system, also based on deep learning. Unfortunately behind a paywall.
  Pinto, J.P., Pimenta, A. & Novais, P. Deep learning and multivariate time series for cheat detection in video games. Mach Learn (2021). https://doi.org/10.1007/s10994-021-06055-x

I think the road ahead will be very rough and any material will probably be interesting to the developers.