mlsanityData to 14 Oct 2026

Are Logistic Models Really Interpretable?

arXiv 2406.13427 ↗First posted 19 Jun 2024cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

The demand for open and trustworthy AI models points towards widespread publishing of model weights. Consumers of these model weights must be able to act accordingly with the information provided. That said, one of the simplest AI classification models, Logistic Regression (LR), has an unwieldy interpretation of its model weights, with greater difficulties when extending LR to generalised additive models. In this work, we show via a User Study that skilled participants are unable to reliably reproduce the action of small LR models given the trained parameters. As an antidote to this, we define Linearised Additive Models (LAMs), an optimal piecewise linear approximation that augments any trained additive model equipped with a sigmoid link function, requiring no retraining. We argue that LAMs are more interpretable than logistic models -- survey participants are shown to solve model reasoning tasks with LAMs much more accurately than with LR given the same information. Furthermore, we show that LAMs do not suffer from large performance penalties in terms of ROC-AUC and calibration with respect to their logistic counterparts on a broad suite of public financial modelling data.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 20 of its 70 references that are in the map. First posted on arXiv 19 Jun 2024.

Builds on

20 of 70 references.

References · 20

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  2. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  3. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  4. 2018Manipulating and Measuring Model Interpretabilitycited by 256
  5. 2019Explainable Machine Learning in Deploymentcited by 211
  6. 2015Should we really use post-hoc tests based on mean-ranks?cited by 116
  7. 2021Do Feature Attribution Methods Correctly Attribute Features?cited by 98
  8. 2018Boolean Decision Rules via Column Generationcited by 80
  9. 2018How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanationcited by 72
  10. 2018Fairness and Accountability Design Needs for Algorithmic Support in High-Stakes Public Sector Decision-Makingcited by 67
  11. 2020Born-Again Tree Ensemblescited by 38
  12. 2018Explainable Neural Networks based on Additive Index Modelscited by 33
8 more
  1. 2019Enhancing Explainability of Neural Networks through Architecture Constraintscited by 29
  2. 2020Enforcing Interpretability and its Statistical Impacts: Trade-offs between Accuracy and Interpretabilitycited by 25
  3. 2021Designing Inherently Interpretable Machine Learning Modelscited by 23
  4. 2021A Holistic Approach to Interpretability in Financial Lending: Models, Visualizations, and Summary-Explanationscited by 16
  5. 2019Forecasting remaining useful life: Interpretable deep learning approach via variational Bayesian inferencescited by 12
  6. 2019Sample size calculations for the experimental comparison of multiple algorithms on multiple problem instancescited by 5
  7. 2022Marginal Effects for Non-Linear Prediction Functionscited by 5
  8. 2023Regionally Additive Models: Explainable-by-design models minimizing feature interactionscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Empirical Characterization of Rationale Stability Under Controlled Perturbations for Explainable Pattern Recognitioncited by 0