Open AccessPosted Content
Supersparse Linear Integer Models for Interpretable Classification
TL;DR: An off-the-shelf tool to create scoring systems that both accurate and interpretable, known as a Supersparse Linear Integer Model (SLIM), which is a discrete optimization problem that minimizes the 0-1 loss to encourage a high level of accuracy.
read more
Abstract: Scoring systems are classification models that only require users to add, subtract and multiply a few meaningful numbers to make a prediction. These models are often used because they are practical and interpretable. In this paper, we introduce an off-the-shelf tool to create scoring systems that both accurate and interpretable, known as a Supersparse Linear Integer Model (SLIM). SLIM is a discrete optimization problem that minimizes the 0-1 loss to encourage a high level of accuracy, regularizes the L0-norm to encourage a high level of sparsity, and constrains coefficients to a set of interpretable values. We illustrate the practical and interpretable nature of SLIM scoring systems through applications in medicine and criminology, and show that they are are accurate and sparse in comparison to state-of-the-art classification models using numerical experiments.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Figures
Fig. 2: Decision tree induced by the SLIM scoring system for the mammo dataset. 
Fig. 5: Sparsity and accuracy of SLIM vs. models on the regularization path of LARS Lasso 
Fig. 7: Computational performance over time for haberman. 
Fig. 6: Computational performance over time for breastcancer. 
Fig. 11: Computational performance over time for tictactoe. 
Fig. 1: SLIM scoring system for the breastcancer dataset.
Citations
•Posted Content
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)
Been Kim,Martin Wattenberg,Justin Gilmer,Carrie J. Cai,James Wexler,Fernanda B. Viégas,Rory Sayres +6 more
TL;DR: In this article, the authors introduce Concept Activation Vectors (CAVs), which provide an interpretation of a neural net's internal state in terms of human-friendly concepts, and use CAVs as part of a technique, Testing with CAVs (TCAV), that uses directional derivatives to quantify the degree to which a user defined concept is important to a classification result.
1.1K
Algorithmic Transparency via Quantitative Input Influence: Theory and Experiments with Learning Systems
Anupam Datta,Shayak Sen,Yair Zick +2 more
- 22 May 2016
TL;DR: The transparency-privacy tradeoff is explored and it is proved that a number of useful transparency reports can be made differentially private with very little addition of noise.
Auditing black-box models for indirect influence
Philip Adler,Casey Falk,Sorelle A. Friedler,Tionney Nix,Gabriel Rybeck,Carlos Scheidegger,Brandon Smith,Suresh Venkatasubramanian +7 more
TL;DR: In this article, the authors present a technique for auditing black-box models, which lets them study the extent to which existing models take advantage of particular features in the data set, without knowing how the models work.
•Posted Content
Explainable Artificial Intelligence: a Systematic Review
Giulia Vilone,Luca Longo +1 more
TL;DR: This systematic review contributes to the body of knowledge by clustering these methods with a hierarchical classification system with four main clusters: review articles, theories and notions, methods and their evaluation.
202
Legal requirements on explainability in machine learning
TL;DR: The increasing number of legal requirements on machine learning model interpretability and explainability in the context of private and public decision making are presented and how those legal requirements can be implemented into machine-learning models are explained.
151
References
Random Forests
Leo Breiman
- 01 Oct 2001
TL;DR: Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the forest, and are also applicable to regression.
Regression Shrinkage and Selection via the Lasso
TL;DR: A new method for estimation in linear models called the lasso, which minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant, is proposed.
Estimating the Dimension of a Model
TL;DR: In this paper, the problem of selecting one of a number of models of different dimensions is treated by finding its Bayes solution, and evaluating the leading terms of its asymptotic expansion.
Estimating the dimension of a model
Gideon Schwarz
- 01 Jan 2005
TL;DR: In this paper, the problem of selecting one of a number of models of different dimensions is treated by finding its Bayes solution, and evaluating the leading terms of its asymptotic expansion.
40.6K
Statistical learning theory
Vladimir Vapnik
- 01 Jan 1998
TL;DR: Presenting a method for determining the necessary and sufficient conditions for consistency of learning process, the author covers function estimates from small data pools, applying these estimations to real-life problems, and much more.
30.4K