Explainability
Properties
Key quality characteristics of the explanation
9 subcategories • 46 techniques
Causality
Identifies causal rather than correlational relationships
Completeness
Attributions fully account for model output
Comprehensibility
Produces human-understandable formats
Consistency
Similar inputs produce similar explanations
Counterfactual Validity
Can show what changes would alter outcomes
Decision Support
Measures whether the explanation improves, or at least preserves, the decisions people make with the model
Efficiency
Computationally efficient to generate
Fidelity
Accurately represents true model behaviour
Sparsity
Focuses on few, most important factors
Techniques
46 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| SHapley Additive exPlanations | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | SHAP (SHapley Additive exPlanations) explains one prediction by sharing out the gap between that prediction and a... | ||
| Mean Decrease Impurity | Algorithmic | Architecture/tree Based Paradigm/supervised +1 | Tabular | Mean Decrease Impurity (MDI) scores how much each feature contributes to a tree-based model, such as a random forest, by... | ||
| Integrated Gradients | Algorithmic | Architecture/neural Networks Architecture/neural Networks/transformer +4 | Image Tabular Text | Integrated Gradients assigns input features scores for a prediction relative to a chosen reference input, called the... | ||
| DeepLIFT | Algorithmic | Requirements/model Internals Architecture/neural Networks +2 | Any | DeepLIFT (Deep Learning Important FeaTures) explains a single prediction of a neural network by dividing the difference... | ||
| Layer-wise Relevance Propagation | Algorithmic | Requirements/model Internals Architecture/neural Networks +3 | Any | Layer-wise Relevance Propagation (LRP) explains a neural network's prediction by working backwards from the output to... | ||
| Contextual Decomposition | Algorithmic | Architecture/neural Networks/recurrent Requirements/white Box +1 | Text | Contextual Decomposition explains why an LSTM, a neural network that reads text one word at a time, made a particular... | ||
| Taylor Decomposition | Algorithmic | Requirements/model Internals Architecture/neural Networks +3 | Any | Taylor Decomposition explains one prediction of a neural network by splitting its output score into a contribution from... | ||
| Sobol Indices | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | Sobol indices show how much of the variation in a model's output can be traced to each of its inputs. The analyst sets a... | ||
| TimeSHAP | Algorithmic | Architecture/neural Networks/recurrent Requirements/black Box | Time-series | TimeSHAP explains one prediction of a model that reads a sequence of events, such as a recurrent neural network that... | ||
| Permutation Importance | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | Permutation Importance measures how much a trained model relies on each input feature. For each feature, the values in a... | ||
| Permutation Tests | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | A permutation test asks how likely an observed result would be if there were no real relationship in the data. The... | ||
| Explanation Max-Sensitivity | Metric | Architecture/model Agnostic Requirements/black Box | Image Tabular | Explanation max-sensitivity measures how stable an explanation is: whether inputs that differ only slightly, in ways... | ||
| Coefficient Magnitudes (in Linear Models) | Metric | Architecture/linear Models Paradigm/parametric +2 | Tabular | Coefficient Magnitudes explains a linear model by reading its fitted coefficients. Each coefficient shows how much the... | ||
| Local Interpretable Model-Agnostic Explanations | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | LIME (Local Interpretable Model-agnostic Explanations) explains individual predictions by approximating the complex... | ||
| Ridge Regression Surrogates | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | A ridge regression surrogate approximates a complex model with a linear equation fitted to the complex model's own... | ||
| Partial Dependence Plots | Visualisation | Architecture/model Agnostic Requirements/black Box | Tabular | Partial Dependence Plots show how changing one or two features affects a model's predictions on average. The technique... | ||
| Individual Conditional Expectation Plots | Visualisation | Architecture/model Agnostic Requirements/black Box | Tabular | Individual Conditional Expectation (ICE) plots display the predicted output for individual instances as a function of a... | ||
| UMAP | Algorithmic | Architecture/model Agnostic Requirements/black Box | Any | UMAP (Uniform Manifold Approximation and Projection) is a non-linear dimensionality reduction method that turns... | ||
| 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.... | ||
| ANCHOR | Algorithmic | Architecture/model Agnostic Requirements/black Box | Tabular | ANCHOR explains one prediction of a black-box classifier with an if-then rule, called an anchor, such as 'age > 30 AND... |
Rows per page
Page 1 of 3