LIME: Explaining Single Predictions from Any ML Model
LIME explains a single prediction from any 'black box' model by approximating it with a simpler model that's only accurate locally. Use it to see why a specific user churned.
LIME explains a single prediction from any 'black box' model by building a simple, interpretable model that's only accurate locally around that point. It's model-agnostic, so you can probe why an XGBoost model flagged one transaction as fraud. LIME works by perturbing the input, getting new predictions, and fitting a weighted linear model to that local data. The main footgun is that the explanation's stability depends heavily on user-defined parameters like neighborhood size and the number of features in the simple model.
Read the original → christophm.github.io
- #mlops
- #explainability
- #machine learning
- #lime
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.