Glossario IA
Il dizionario completo dell'Intelligenza Artificiale
Model-agnostic vs Model-specific
Distinction between explanation techniques that can be applied to any type of model (model-agnostic) and those designed specifically for certain architectures (model-specific). Model-agnostic approaches offer increased flexibility while specific methods can provide more precise explanations.
Post-hoc Explanations
Interpretation methods applied after training a complex model without modifying its structure or performance. These techniques contrast with intrinsic interpretability where the model is designed from the outset to be interpretable.
Intrinsic Interpretability
Property of models that are naturally understandable by design, such as linear regressions or shallow decision trees. These models do not need post-hoc methods to be explained because their internal structure is already transparent.