0% completed

All LessonsFree Lessons (3)

Introduction to Explainable AI

Course Overview

Primer on Gradients

Coding Exercise on Derivatives

Introductory Quiz on Explainable AI

Project

Explainable AI: The Power of Interpreting ML Models

Saliency Maps

Introduction to Saliency Maps—The Vanilla Gradient Guided Backpropagation Smooth Gradient Saliency Integrated Gradients LIME Layer-wise Relevance Propagation LRP-γ and LRP-Ɛ Rules SHAP Coding Exercise on Saliency Maps Quiz on Saliency Maps

Class Activation Maps

Introduction to Class Activation Maps GradCAM X-GradCAM Eigen-CAM Coding Exercise on Class Activation Maps Quiz on Class Activation Maps

Miscellaneous Methods

Counterfactual Explanations Perturbation-Based Explanations Concept Explanations Prototypical Explanations Coding Exercise on Miscellaneous Methods Quiz on Miscellaneous Methods

Metrics of Interpretability

Ground-Truth Faithfulness—Feature Agreement Ground-Truth Faithfulness—Rank Correlation Predictive Faithfulness Stability and Fairness Coding Exercise on Metrics of Interpretability Quiz on Metrics of Interpretability Course Conclusion

Course Assessment

Final Assessment

Mini Project

Weakly Supervised Object Localization

Master Explainable AI: Interpreting Image Classifier Decisions/

...

/

Primer on Gradients

Primer on Gradients

Learn about gradients and PyTorch's autograd API.

We'll cover the following...

Implementing explanation methods requires basic knowledge of gradients. So before diving deeper into the course, let’s first quickly review the concepts of gradients.

Gradients

Given a function $f(x)$ , where $x = (x_1, x_2, ... x_d)$ is a $d$ -dimensional vector, the gradient of $f(x)$ with respect to $x$ , i.e., $\nabla_x f(x)$ , is a ...