Interpretable Credit Application Predictions With Counterfactual Explanations

By Rory Mc Grath, Luca Costabello, Chan Le Van, Paul Sweeney, F. Kamiab, Zhao Shen, Freddy Lécué. 2018

122 citations recorded by the source corpus.

We predict credit applications with off-the-shelf, interchangeable black-box classifiers and we explain single predictions with counterfactual explanations. Counterfactual explanations expose the minimal changes required on the input data to obtain a different result e.g., approved vs rejected application. Despite their effectiveness, counterfactuals are mainly designed for changing an undesired outcome of a prediction i.e. loan rejected. Counterfactuals, however, can be difficult to interpret, especially when a high number of features are involved in the explanation. Our contribution is two-fold: i) we propose positive counterfactuals, i.e. we adapt counterfactual explanations to also explain accepted loan applications, and ii) we propose two weighting strategies to generate more interpretable counterfactuals. Experiments on the HELOC loan applications dataset show that our contribution outperforms the baseline counterfactual generation strategy, by leading to smaller and hence more interpretable counterfactuals.

Subfields: Accounting, Finance, Management Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Accounting

1. Isabel Valera
2. Gilles Barthe
3. Amir-Hossein Karimi
4. Been Kim

In Finance

1. Maarten de Rijke
2. Bernhard Schölkopf

In Management Information Systems

1. Andrew Smart
2. Bernhard Schölkopf
3. Atoosa Kasirzadeh

Citations in the corpus

• Causal Interpretability for Machine Learning - Problems, Methods and Evaluation (2020, methodology): “Traditional Interpretability Interpretable Models: [106], [103], [64], [102], [65], [105], [50], [13], [40] Causal Interpretability Model-based: [77], [38], [12], [8], [112], [81], [7], Example-based: [35], [45], [39], [101], [76], [86], [73], [61], [60], [33], [34] Post-hoc: [47], [89], [66], [93],”
• A survey of algorithmic recourse: definitions, formulations, solutions, and prospects (2020, background): “11) Weighted CFs [71] E heuristic query (2019.”
• FOCUS: Flexible Optimizable Counterfactual Explanations for Tree Ensembles (2019, methodology): “Counterfactual examples have been used in a variety of ML areas, such as reinforcement learning [23], deep learning [1], and explainable AI (XAI) [8, 13, 17, 18, 22, 24, 26, 28, 31, 35, 36].”
• The Use and Misuse of Counterfactuals in Ethical Machine Learning (2021, background): “The use of counterfactuals has become increasingly popular in the machine learning community for many reasons such as making sense of algorithmic fairness or explainability in automated decision-making for consequential social contexts [4, 9, 14, 18, 29, 35, 42, 47, 51].”
• Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems (2019, background): “For instance, if a credit loan is approved, Grath et al. (2018) show the margin (for each client attribute) by which the loan was approved and also provide a margin by which a loan is denied.”
• Counterfactual Explanations for Machine Learning: A Review (2020, methodology): “Black-box approaches use gradient-free optimization algorithms such as Nelder-Mead [56], growing spheres [74], FISTA [30, 108], or genetic algorithms [72, 98, 26] to solve the optimization problem.”
• Multi-Objective Counterfactual Explanations (2020, methodology): “ral principles which work for arbitrary ML models - often by only assuming access to the prediction function of an already tted model. Several model-agnostic counterfactual methods have been proposed [8,11,16,18,25,29,37]. Apart from Grath et al. [11], these approaches are limited to classication. U”
• The hidden assumptions behind counterfactual explanations and principal reasons (2019, background): “, the literature on counterfactual explanations has mostly converged on a heuristic that finds the Median Absolute Deviation (MAD) under an L1 distance norm [15, 29, 36].”
• Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review (2020, unknown): “[124] Black-box Agnostic No No No No No No1 -”