Covers: theory of Explainable AI
Estimated time needed to finish: 5 minutes
Questions this item addresses:
  • Where can i read more about explainable AI?
0 comment

Explainable AI

Total time needed: ~46 minutes
This recipe is a work in progress with the goal of providing curated and hands on resources for various aspects of explainable AI.
Potential Use Cases
blackbox model interpretation, model risk assessment, model bias reduction
Who is This For ?
INTERMEDIATEData Scientists, Machine Learning Engineers
Click on each of the following annotated items to see details.
Resource Asset5/14
WRITEUP 1. Explainable AI Introduction
  • What is this recipe and how can i use it?
5 minutes
VIDEO 2. The Need for Machine Learning Explanations
  • Why Do We Need Machine Learning Explanations?
5 minutes
VIDEO 3. Explainability Options
  • What are various ways we can explain ML models?
6 minutes
RECIPE 4. Local interpretable model-agnostic explanations
10 minutes
RECIPE 5. SHAP (SHapley Additive exPlanations)
10 minutes
WRITEUP 6. XAI Reading List
  • Where can i read more about explainable AI?
5 minutes
  • What tools can I use for explainable AI?
5 minutes
BOOK_CHAPTER 8. Interpretability through Model-Agnostic Methods
  • What options do I have to add interpratbility to existing models?
20 minutes
VIDEO 9. Explainable AI with Layer-wise Relevance Propagation (LRP)
10 minutes
VIDEO 10. Explaining image classifiers by removing input features using generative models
10 minutes
VIDEO 11. Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
10 minutes
VIDEO 12. TF-Encrypted: Private machine learning in tensorflow with secure computing
10 minutes
VIDEO 13. Explaining by Removing: A Unified Framework for Model Explanation
10 minutes
VIDEO 14. High-frequency Component Helps Explain the Generalization of Convolutional Neural Networks
10 minutes

Concepts Covered

Paul Ntalo.
Thank you for providing such curated study resources