\( \newcommand{\vect}[1]{\mathbf{#1}} \newcommand{\mat}[1]{\mathbf{#1}} \newcommand{\R}{\mathbb{R}} \newcommand{\E}{\mathbb{E}} \newcommand{\Prob}{\mathbb{P}} \newcommand{\T}{^{\top}} \newcommand{\norm}[1]{\lVert #1 \rVert} \newcommand{\abs}[1]{\lvert #1 \rvert} \newcommand{\ihat}{\mkern1mu\hat{\boldsymbol{\imath}}\mkern2mu} \newcommand{\jhat}{\mkern1mu\hat{\boldsymbol{\jmath}}\mkern2mu} \newcommand{\pd}[2]{\frac{\partial #1}{\partial #2}} \newcommand{\argmax}{\operatorname*{arg\,max}} \newcommand{\argmin}{\operatorname*{arg\,min}} \newcommand{\cdata}[1]{\class{c-data}{#1}} \newcommand{\cmodel}[1]{\class{c-model}{#1}} \newcommand{\ctruth}[1]{\class{c-truth}{#1}} \newcommand{\cgrad}[1]{\class{c-grad}{#1}} \newcommand{\closs}[1]{\class{c-loss}{#1}} \newcommand{\cparam}[1]{\class{c-param}{#1}} \newcommand{\csecond}[1]{\class{c-second}{#1}} \renewcommand{\,}{\mkern3mu} \)

R02 · Interpretability and Explainability

A free, illustrated L3 course of 9 modules in the Responsible AI track. Modules include: What an explanation is; Interpretable models; Feature attribution: permutation, SHAP, LIME.

R02 L3 · Advanced 9 modules · about 36 lessons Planned · Ring 3

Part of the Responsible AI track. Ships in Ring 3 (Breadth) of the roadmap.

Modules

1. What an explanation is
2. Interpretable models
3. Feature attribution: permutation, SHAP, LIME
4. Saliency and gradient methods
5. Concept-based methods
6. Probing
7. Mechanistic interpretability: features and circuits
8. Sparse autoencoders
9. Evaluating explanations

Signature illustrations

The pictures a learner should still see in their head a year later:

Anatomy of an attribution waterfall, saliency overlays, circuit diagrams, galleries of feature activations.

Status

This course is planned, not written yet. It ships with Ring 3 (Breadth).

Every lesson will follow the same twelve-part blueprint with at least three purpose-made figures. Watch the repository to hear when modules ship, or help write and review it.