Generalized Additive Models

Description

An intrinsically interpretable modelling technique that extends linear models by allowing flexible, nonlinear relationships between individual features and the target whilst maintaining the additive structure that preserves transparency. Each feature's effect is modelled separately as a smooth function, visualised as a curve showing how the feature influences predictions across its range. GAMs achieve this through spline functions or other smoothing techniques that capture complex patterns in individual variables without interactions, making them particularly valuable for domains requiring both predictive accuracy and model interpretability.

Example Use Cases

Transparency

Predicting hospital readmission risk with a GAM that provides transparent, auditable risk assessments by showing how readmission probability varies nonlinearly with patient age, blood pressure, and medication adherence, enabling clinicians to understand and trust the model's reasoning for regulatory compliance.

Explainability

Building a credit scoring model that explains loan decisions to applicants by visualising how income, credit history, and debt-to-income ratio individually affect approval likelihood, providing clear feature attributions that satisfy fair lending requirements and regulatory explainability mandates.

Reliability

Developing an environmental monitoring system that reliably predicts air quality using GAMs to model the smooth, nonlinear relationships between weather variables, ensuring stable predictions across seasonal variations whilst maintaining interpretable relationships that environmental scientists can validate.

Limitations

Cannot capture complex interactions between features unless explicitly modelled, limiting their ability to represent relationships where variables influence each other.
Setup requires domain expertise to decide which features need nonlinear treatment and appropriate smoothing parameters, making model specification more challenging than linear models.
Fitting process is computationally more expensive than linear models, particularly for large datasets with many features requiring smoothing.
Risk of overfitting individual feature relationships if smoothing parameters are not properly regularised, potentially reducing generalisation performance.
Interpretation complexity increases with the number of nonlinear features, as understanding multiple smooth curves simultaneously becomes cognitively demanding.

Resources

Generalized Additive Models

Research Paper•Trevor Hastie and Robert Tibshirani•Jan 1, 1986

pyGAM: Generalized Additive Models in Python

Software Package

mgcv: Mixed GAM Computation Vehicle with Automatic Smoothness Estimation

Software Package

A Tour of pyGAM — pyGAM documentation

Tutorial

Related Techniques

Name	Description	Assurance Goals
Attribute Removal (Fairness Through Unawareness)	Attribute Removal (Fairness Through Unawareness) ensures fairness by completely excluding protected attributes such as race, gender, or age from the model's input features. While this approach prevents direct discrimination, it may not eliminate bias if other features are correlated with protected attributes (proxy discrimination). This technique represents the most basic fairness intervention but often needs to be combined with other approaches to address indirect bias through seemingly neutral features.	Fairness Transparency
Contrastive Explanation Method	The Contrastive Explanation Method (CEM) explains model decisions by generating contrastive examples that reveal what makes a prediction distinctive. It identifies 'pertinent negatives' (minimal features that could be removed to change the prediction) and 'pertinent positives' (minimal features that must be present to maintain the prediction). This approach helps users understand not just what led to a decision, but what would need to change to achieve a different outcome, providing actionable insights for decision-making.	Explainability Transparency
Coefficient Magnitudes (in Linear Models)	Coefficient Magnitudes assess feature influence in linear models by examining the absolute values of their coefficients. Features with larger absolute coefficients are considered to have a stronger impact on the prediction, while the sign of the coefficient indicates the direction of that influence (positive or negative). This technique provides a straightforward and transparent way to understand the direct linear relationship between each input feature and the model's output.	Explainability Transparency
ANCHOR	ANCHOR generates high-precision if-then rules that explain individual predictions by identifying the minimal set of feature conditions that guarantee a specific prediction with high confidence. It searches for 'anchor' conditions (e.g., 'age > 30 AND income < £50k') that ensure the model gives the same prediction at least 95% of the time when those conditions are met. This creates human-readable rules that users can trust as sufficient conditions for understanding why a particular decision was made.	Explainability Transparency
Empirical Calibration	Empirical calibration adjusts a model's predicted probabilities to match observed frequencies. For example, if events predicted with 80% confidence only occur 60% of the time, calibration would correct this overconfidence. Common techniques include Platt scaling and isotonic regression, which learn transformations that map the model's raw scores to well-calibrated probabilities, improving the reliability of confidence measures for downstream decisions.	Reliability Transparency Fairness
Prompt Sensitivity Analysis	Prompt Sensitivity Analysis systematically evaluates how variations in input prompts affect large language model outputs, providing insights into model robustness, consistency, and interpretability. This technique involves creating controlled perturbations of prompts whilst maintaining semantic meaning, then measuring how these changes influence model responses. It encompasses various types of prompt modifications including lexical substitutions, syntactic restructuring, formatting changes, and contextual variations. The analysis typically quantifies sensitivity through metrics such as output consistency, semantic similarity, and statistical measures of variance across prompt variations.	Explainability Reliability Safety

Generalized Additive Models

Description

Example Use Cases

Transparency

Explainability

Reliability

Limitations

Resources

Generalized Additive Models

pyGAM: Generalized Additive Models in Python

mgcv: Mixed GAM Computation Vehicle with Automatic Smoothness Estimation

A Tour of pyGAM — pyGAM documentation

Related Techniques

Tags