Hide Navigation Hide TOC

Adversarial Input Detection (0ed2ef71-cdc9-4eef-8432-1c3dadbdda20)

Detect and block adversarial inputs or atypical queries that deviate from known benign behavior, exhibit behavior patterns observed in previous attacks or that come from potentially malicious IPs. Incorporate adversarial detection algorithms into the ML system prior to the ML model.

Cluster A	Galaxy A	Cluster B	Galaxy B	Level
Denial of ML Service (8f644f37-e2e6-468e-b720-f395b8c27fbc)	MITRE ATLAS Attack Pattern	Adversarial Input Detection (0ed2ef71-cdc9-4eef-8432-1c3dadbdda20)	MITRE ATLAS Course of Action	1
Black-Box Optimization (c4e52005-7416-45c4-9feb-8cd5fd34f70a)	MITRE ATLAS Attack Pattern	Adversarial Input Detection (0ed2ef71-cdc9-4eef-8432-1c3dadbdda20)	MITRE ATLAS Course of Action	1
Erode ML Model Integrity (8735735d-c09d-4298-8e64-9a2b6168a74c)	MITRE ATLAS Attack Pattern	Adversarial Input Detection (0ed2ef71-cdc9-4eef-8432-1c3dadbdda20)	MITRE ATLAS Course of Action	1
Evade ML Model (071df654-813a-4708-85dc-f715f785d37f)	MITRE ATLAS Attack Pattern	Adversarial Input Detection (0ed2ef71-cdc9-4eef-8432-1c3dadbdda20)	MITRE ATLAS Course of Action	1
Black-Box Optimization (c4e52005-7416-45c4-9feb-8cd5fd34f70a)	MITRE ATLAS Attack Pattern	Craft Adversarial Data (a7c30122-b393-4265-91b7-57cd1211e3f9)	MITRE ATLAS Attack Pattern	2