Intrinsically Interpretable Models

Description

Intrinsically interpretable models are machine learning algorithms that are transparent by design, allowing users to understand their decision-making process without requiring additional explanation techniques. This category includes decision trees and rule lists (which use if-then logic), linear and logistic regression models (which use weighted feature combinations), and other simple algorithms where the model structure itself provides interpretability. These models prioritise transparency over complexity, making them ideal when stakeholder understanding and regulatory compliance are paramount.

Example Use Cases

Transparency

Developing a medical diagnosis support system using a decision tree with clear if-then rules based on symptoms and test results, allowing healthcare professionals to trace the reasoning path and explain diagnoses to patients whilst ensuring clinical transparency and accountability.

Implementing a hiring decision support tool using rule lists that explicitly state qualification criteria and scoring logic, providing transparent candidate evaluation that can be explained to applicants and reviewed for fairness whilst meeting legal requirements for employment decision documentation.

Reliability

Creating a fraud detection model using logistic regression with carefully selected features (transaction amount, location, time patterns) where each coefficient's contribution can be understood and validated, ensuring reliable performance that financial institutions can audit and regulatory bodies can approve.

Limitations

Generally achieve lower predictive accuracy than complex models (neural networks, ensembles) for difficult problems involving high-dimensional data, non-linear relationships, or complex feature interactions.
Linear models cannot capture non-linear relationships or feature interactions without manual feature engineering, limiting their applicability to inherently non-linear domains like image recognition or natural language processing.
Decision trees can become unstable with small changes in training data, potentially leading to completely different tree structures and predictions, affecting model reliability in dynamic environments.
Deep decision trees may lose interpretability despite being inherently transparent, as human cognitive limits make it difficult to follow complex branching logic with many levels and conditions.
Feature selection becomes critical for maintaining interpretability, requiring domain expertise to identify the most relevant variables whilst potentially missing important but subtle predictive signals.

Resources

scikit-learn Decision Trees

Documentation

Comprehensive documentation for decision tree implementation in scikit-learn, including classification and regression trees with interpretability guidelines and visualisation tools.

scikit-learn Linear Models

Documentation

Complete guide to linear and logistic regression models in scikit-learn, covering implementation, feature selection, and coefficient interpretation for transparent modeling.

Interpretable Machine Learning

Tutorial

Open-source book providing comprehensive coverage of interpretable machine learning models including decision trees, linear models, and rule-based systems with practical examples.

R package 'rpart' for Recursive Partitioning

Software Package

R implementation of recursive partitioning for classification, regression and survival trees with extensive documentation and plotting capabilities for interpretable tree models.

Related Techniques

Name	Description	Assurance Goals
MLflow Experiment Tracking	MLflow is an open-source platform that tracks machine learning experiments by automatically logging parameters, metrics, models, and artifacts throughout the ML lifecycle. It provides a centralised repository for comparing different experimental runs, reproducing results, and managing model versions. Teams can track hyperparameters, evaluation metrics, model files, and execution environment details, creating a comprehensive audit trail that supports collaboration, reproducibility, and regulatory compliance across the entire machine learning development process.	Transparency Reliability
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
Internal Review Boards	Internal Review Boards (IRBs) provide independent, systematic evaluation of AI/ML projects throughout their lifecycle to identify ethical, safety, and societal risks before they materialise. Typically composed of multidisciplinary experts including ethicists, domain specialists, legal counsel, community representatives, and technical staff, IRBs review project proposals, assess potential harms to individuals and communities, evaluate mitigation strategies, and establish ongoing monitoring requirements. Unlike traditional research ethics committees, AI-focused IRBs address algorithmic bias, fairness concerns, privacy implications, and societal impact at scale, providing essential governance for responsible AI development and deployment.	Safety Fairness Transparency
Bootstrapping	Bootstrapping estimates uncertainty by repeatedly resampling the original dataset with replacement to create many new training sets, training a model on each sample, and analysing the variation in predictions. This approach provides confidence intervals and stability measures without making strong statistical assumptions. By showing how predictions change with different random samples of the data, it reveals how sensitive the model is to the specific training examples and provides robust uncertainty estimates.	Reliability Transparency Fairness
Homomorphic Encryption	Homomorphic encryption allows computation on encrypted data without decrypting it first, producing encrypted results that, when decrypted, match the results of performing the same operations on the plaintext. This enables secure outsourced computation where sensitive data remains encrypted throughout processing. By allowing ML operations on encrypted data, it provides strong privacy guarantees for applications involving highly sensitive information.	Privacy Safety Transparency Security
Conformal Prediction	Conformal prediction provides mathematically guaranteed uncertainty quantification by creating prediction sets that contain the true outcome with a specified probability (e.g., exactly 95% coverage). The technique works by measuring how 'strange' or 'nonconforming' new predictions are compared to calibration data - if a prediction seems unusual, it gets wider intervals. For example, in medical diagnosis, instead of saying 'likely cancer', it might say 'possible diagnoses: {cancer, benign tumour} with 95% confidence'. This distribution-free method works with any underlying model (neural networks, random forests, etc.) and requires no assumptions about data distribution, making it a robust framework for reliable uncertainty estimates in high-stakes applications.	Reliability Transparency Fairness