MRG Effitas supports Machine Learning Security Evasion Competition

MRG Effitas is pleased once again to support the annual machine learning (ML) security evasion competition with the use of its intellectual property and technology.

The competition, a collaboration between Microsoft, MRG Effitas, CUJO AI, VMRay and NVIDIA, encourages researchers to compete against their peers and exercise their defender and attacker muscles against ML security models in a unique real-world setting.

The contest is designed to spotlight countermeasures to malicious behaviour and raise awareness of the various ways in which threat actors can evade ML systems.

The Defender Challenge ran from June 15 through July 23, 2021, during which contestants submitted malware detection models, which then became targets for evasion for the Attacker Challenge. To win the Defender Challenge, the models must allow the fewest evasions during the subsequent Attacker Challenge round.

The Attacker Challenge is running from August 6 through September 17, 2021, and this year involves two challenges: an Anti-Malware Evasion track and an Anti-Phishing Evasion track.

Anti-Phishing Evasion Track: Contestants play the role of an attacker and attempt to evade a suite of anti-phishing models. Custom built by CUJO AI, the phishing machine learning models are purpose-built for this competition only.
Anti-Malware Evasion track: This challenge provides an alternative scenario for attackers wishing to bypass machine-learning-based antivirus: change an existing malicious binary in a way that disguises it from the antimalware model.
In addition, for each of the Attacker Challenge tracks, the highest-scoring submission that leverages Counterfit —Microsoft’s open-source tool for investigating the security of machine learning models—will be awarded a bonus prize.

“The competition is a great way to support Machine Learning research, improve AI-based malware detection and help talented young researchers to make a name for themselves, so it’s great to be a part of it,” says Chris Pickard, CEO of MRG Effitas.

It’s not too late to participate in the Attacker Challenge which runs til 17th September. Individuals or teams can register at https://mlsec.io.

Prizes will be awarded for each challenge and to win, researchers must publish their detection or evasion strategies. The winners will be announced at the beginning of October.