Model-agnostic explanations

Explainable Artificial Intelligence (XAI) Concepts

Folkert Stijnman

ML Engineer

Black box models

an illustration depicting a magician's hat

  • We know what goes in
  • We know what comes out
  • In between are a lot of internal parameters
Explainable Artificial Intelligence (XAI) Concepts

Global and local explanations

 

  • SHAP (SHapley Additive exPlanations)
  • LIME (Local Interpretable Model-agnostic Explanations)

 

a picture showing a transparent sphere

Explainable Artificial Intelligence (XAI) Concepts

SHAP (SHapley Additive exPlanations)

an illustration depicting a detective

an illustration depicting a cake

 

  • Acts like a detective
  • Look at each feature
  • Helps to figure out the relative impact of each ingredient on the final outcome
Explainable Artificial Intelligence (XAI) Concepts

Shapley values

  • Concept from cooperative game theory
  • Assess impact of features
  • Which features matter for the model's prediction

SHAP example

Explainable Artificial Intelligence (XAI) Concepts

SHAP image recognition

SHAP example

Explainable Artificial Intelligence (XAI) Concepts

LIME (Local Interpretable Model-agnostic Explanations)

illustration of scientist in a lab setting

  • Acts like a scientist
  • Change input to see outcome
  • What if we change some pixels in this image?
Explainable Artificial Intelligence (XAI) Concepts

Building a self-driving car

SHAP

  • Global understanding
  • Impact of car features on overall performance

LIME

  • Local understanding
  • Impact of car features on a specific decision
Explainable Artificial Intelligence (XAI) Concepts

Let's practice!

Explainable Artificial Intelligence (XAI) Concepts

Preparing Video For Download...