Input Restoration - AI Mitigation
AI MitigationPreprocess all inference data to nullify or reverse potential adversarial perturbations.
Overview
A source-backed snapshot of this defense.
Techniques8Attacks this defense is designed to help with.
Lifecycle4Where this defense applies in the AI lifecycle.
Categories1How the source groups this defense.
Safeguard details
Where this defense applies and how the source classifies it.
- ATLAS ID
- AML.M0010
- Priority score
- 40
Data PreparationML Model EvaluationDeploymentMonitoring and Maintenance
Technical - ML
Covered techniques
Attacks this defense is designed to help with.
8 recordsView all techniques →
AML.T0043.001 - Black-Box Optimization
Input restoration adds an extra layer of unknowns and randomness when an adversary evaluates the input-output relationship.
demonstrated
AML.T0043.002 - Black-Box Transfer
Input restoration can help remediate adversarial inputs.
demonstrated
AML.T0043 - Craft Adversarial Data
Input restoration can help remediate adversarial inputs.
realized
AML.T0031 - Erode AI Model Integrity
Preprocessing model inputs can prevent malicious data from going through the machine learning pipeline.
realized
Showing 4 of 8
Source evidence
Original public records and references for this page.
Original source
Original source links
Open the public records and source datasets used for this page.