brief-content

brief content

 

    

Part 1. Interpretability basics

  1 Introduction

  2 White-box models

Part 2. Interpreting model processing

  3 Model-agnostic methods: Global interpretability

  4 Model-agnostic methods: Local interpretability

  5 Saliency mapping

Part 3. Interpreting model representations

  6 Understanding layers and units

  7 Understanding semantic similarity

Part 4. Fairness and bias

  8 Fairness and mitigating bias

  9 Path to explainable AI

  

Appendix A. Getting set up

Appendix B. PyTorch