Słownik AI

📖

pojęcia

DeepLIFT (Deep Learning Important FeaTures)

Attribution method for deep neural networks that compares each neuron's activation to a reference to calculate feature contributions.

📖

pojęcia

Local vs Global Explanability

Distinction between explainability that explains an individual prediction (local) and that which describes the general behavior of the model over the entire dataset (global).

📖

pojęcia

Feature Interaction Effects

Measure of synergistic dependence between two or more features where their combined effect on the prediction differs from the sum of their individual effects.