Montimage AI Platform (MAIP) Documentation
  • Overview
  • Architecture
  • Installation
    • Installation instructions
    • Docker
  • MAIP API reference
    • Reports
      • /
      • Download
    • Logs
      • /
      • Download
    • Retrain
      • /
    • Models
      • /
      • App
      • Download
      • Confusion matrix
      • Build config
      • Predictions
      • Probabilities
      • Datasets
        • /
        • Download
    • Predictions
      • /
      • Download
      • Attack
      • Normal
    • Mmt
      • /
      • Stop
      • Offline
      • Dataset
      • Online
    • Pcaps
      • /
      • Datasets
      • Dataset
    • Build
      • /
    • Ac
      • Retrain
        • /
      • Datasets
        • /
      • Build
        • /
    • Predict
      • /
      • Stop
    • Xai
      • /
      • Shap
        • Explanations
      • Lime
        • Explanations
      • Explanations
    • Attacks
      • /
      • Ctgan
        • Download
      • Datasets
      • Poisoning
        • Ctgan
        • Random swapping labels
        • Target label flipping
        • Download
    • Metrics
      • Accuracy
      • Currentness
      • Impact
  • Specification
  • Client side
    • All Models page
      • Models Comparison page
      • Models Retraining page
    • Dataset page
    • Build Page
    • Predict Page
    • XAI page
      • XAI SHAP Page
      • XAI LIME Page
    • Attacks Page
    • Metrics Page
      • Accountability Metrics Page
      • Resilience Metrics Page
  • About
Powered by GitBook
On this page
  1. Client side
  2. Metrics Page

Resilience Metrics Page

This page provides different resilience metrics of a model.

PreviousAccountability Metrics PageNextAbout

Last updated 1 year ago

Impact metric shows difference between the original accuracy of a benign model compared to the accuracy of the compromised model after a successful poisoning attack.