FACE: Feasible and Actionable Counterfactual Explanations

By Rafael Poyiadzi, Kacper Sokol, Raúl Santos-Rodríguez, Tijl De Bie, Peter A. Flach. 2019

478 citations recorded by the source corpus.

Work in Counterfactual Explanations tends to focus on the principle of "the closest possible world" that identifies small changes leading to the desired outcome. In this paper we argue that while this approach might initially seem intuitively appealing it exhibits shortcomings not addressed in the current literature. First, a counterfactual example generated by the state-of-the-art systems is not necessarily representative of the underlying data distribution, and may therefore prescribe unachievable goals (e.g., an unsuccessful life insurance applicant with severe disability may be advised to do more sports). Secondly, the counterfactuals may not be based on a "feasible path" between the current state of the subject and the suggested one, making actionable recourse infeasible (e.g., low-skilled unsuccessful mortgage applicants may be told to double their salary, which may be hard without first increasing their skill level). These two shortcomings may render counterfactual explanations impractical and sometimes outright offensive. To address these two major flaws, first of all, we propose a new line of Counterfactual Explanations research aimed at providing actionable and feasible paths to transform a selected instance into one that meets a certain goal. Secondly, we propose FACE: an algorithmically sound way of uncovering these "feasible paths" based on the shortest path distances defined via density-weighted metrics. Our approach generates counterfactuals that are coherent with the underlying data distribution and supported by the "feasible paths" of change, which are achievable and can be tailored to the problem at hand.

Subfields: Artificial Intelligence, Economics and Econometrics, General Decision Sciences

How researchers see this paper

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

In Artificial Intelligence

In Economics and Econometrics

In General Decision Sciences

Citations in the corpus

Algorithmic recourse under imperfect causal knowledge: a probabilistic approach (2020, background): “5, (2) where dist(·, ·) is a similarity metric on X , and additional constraints may be added to reflect plausibility, feasibility, or diversity of the obtained counterfactual explanations [19, 20, 32, 33, 39, 44].”
Getting a CLUE: A Method for Explaining Uncertainty Estimates (2020, background): “[23] Rafael Poyiadzi, Kacper Sokol, Raul Santos-Rodriguez, Tijl De Bie, and Peter Flach.”
A survey of algorithmic recourse: definitions, formulations, solutions, and prospects (2020, methodology): “…Guided Prototypes [173] E grad opt/FISTA gradient/query (2019.07) REVISE [87] E grad opt gradient (2019.08) CLEAR [182] E heuristic query (2019.08) MC-BRP [113] E heuristic query (2019.09) FACE [149] E graph + heuristic query (2019.10) Action Sequences [150] E program synthesis class prob.”
Multi-Objective Counterfactual Explanations (2020, methodology): “Unlike the other methods, the method of Poyiadzi et al. [23] can obtain plausible counterfactuals by constructing feasible paths between data points with opposite predictions.”
Explainability fact sheets: a framework for systematic assessment of explainable approaches (2019, background): “Since we have not seen many data points in that region, we should not trust the explanation without further investigation [30, 40].”
Algorithmic Recourse: from Counterfactual Explanations to Interventions (2020, background): “…x CFE ∗ ∈ X is a (perhaps not unique) nearest counterfactual instance; h θ is the ﬁxed predictor; and P is an optional set of plausibility constraints, e.g., the counterfactual instance be from a relatively high density region of the input space (Joshi et al., 2019; Poyiadzi et al., 2019).”
A Survey of Contrastive and Counterfactual Explanation Generation Methods for Explainable Artificial Intelligence (2021, background): “However, these terms are sometimes used interchangeably in the context of XAI [48], [49].”
FOCUS: Flexible Optimizable Counterfactual Explanations for Tree Ensembles (2019, methodology): “Fidelity is commonly used to evaluate XAI methods that are based on approximations of a givenmodel: it is ameasure of the agreement between the original modelM and its approximation M̃: fid(M̃, 𝑋 )= 1 𝑁 𝑁∑︁ 𝑛=1 1 [ 𝑦𝑥 (𝑛) =arg max 𝑦′ M̃(𝑦′ |𝑥 (𝑛) ) ] , (26) where 𝑦𝑥 = arg max𝑦 M(𝑦 | �”
Counterfactual Explanations for Machine Learning: A Review (2020, methodology): “Poyiadzi et al. [89] propose FACE, which uses Dijkstra’s algorithm [31] to ﬁnd the shortest path between existing training datapoints to ﬁnd counterfactual for a given input.”
Data-centric Artificial Intelligence: A Survey (2023, unknown): “For example, Dijkstra’s algorithm is employed to obtain the shortest path between existing training data points to find recourse under certain distributions [173].”
Benchmarking and survey of explanation methods for black box models (2021, methodology): “FACE, Feasible and Actionable Counterfactual Explanations (Poyiadzi et al. 2020) is a model-agnostic post-hoc explainer that focuses on returning actionable counterfactuals, i.e., records coherent with the input data distribution.”
Towards Robust and Reliable Algorithmic Recourse (2021, background): “Several approaches in recent literature tackled the problem of providing recourses by generating local (instance level) counterfactual explanations 2 [31, 26, 12, 21, 18].”
Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers (2019, background): “To account for feasibility, different statistical notions of feasibility have been proposed, based on adherence to the training distribution [5], distribution of the target class [14], the likelihood of the intermediate points in reaching a CF example [6], and on specifying which features can be cha”
A Survey on the Explainability of Supervised Machine Learning (2020, background): “The Feasible and Actionable Counterfactual Explanations (FACE) approach aims to build coherent and feasible counterfactuals by using the shortest path distances defined via density-weighted metrics (Poyiadzi et al., 2020).”