Explainability
Instance-Based Methods
Techniques using example-based explanations
3 subcategories • 4 techniques
Counterfactual
Shows alternative scenarios that would change outcome (e.g., Contrastive Explanation Method)
Influence Analysis
Traces impact of training data on predictions (e.g., Influence Functions)
Prototypes
Uses representative examples for explanation (e.g., Prototype & Criticism Models)
Techniques
4 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| Prototype and Criticism Models | Algorithmic | Architecture/model Agnostic Paradigm/supervised +2 | Any | Prototype and criticism methods summarise a dataset with a small set of real examples. Prototypes are instances chosen... | ||
| Influence Functions | Algorithmic | Architecture/linear Models Architecture/neural Networks +5 | Any | Influence functions estimate which training examples most affected a model's prediction on a particular test example.... | ||
| Contrastive Explanation Method | Algorithmic | Architecture/neural Networks Paradigm/discriminative +4 | Any | The Contrastive Explanation Method (CEM) explains one prediction of a classifier, such as a neural network, by... | ||
| Diverse Counterfactual Explanations | Algorithmic | Architecture/model Agnostic Paradigm/supervised +1 | Tabular | Diverse Counterfactual Explanations (DiCE) explains one decision of a classifier by generating several altered versions... |
Rows per page
Page 1 of 1