# Evasion Attack

**IRI:** https://folio.openlegalstandard.org/R8nzGmaooF9VdUO5jELI3Et

## Labels

**Preferred Label:** Adversarial Evasion

**Alternative Labels:**

- Deceptive Input Attack
- Evasion
- Misclassification Attack

## Definition

An evasion attack is one of the most common attacks on machine learning models performed during production. It involves designing an input that appears normal to a human but is misclassified by ML models, affecting their behavior; a typical example includes altering pixels in an image to cause an image recognition system to fail in classification.

## Sub Class Of

- https://folio.openlegalstandard.org/RCafxux1zDpXGZzqVBKNVpa

## See Also

- https://folio.openlegalstandard.org/RBMj5dbvLFgFGrPZ3MvFGYl

**Deprecated:** False