Learning Model-Agnostic Counterfactual Explanations for Tabular Data

Pawelczyk, Martin; Haug, Johannes; Broelemann, Klaus; Kasneci, Gjergji

doi:10.1145/3366423.3380087

Cited by 131 publications

(136 citation statements)

References 10 publications

Supporting

Mentioning

136

Contrasting

Order By: Relevance

“…The explanation generation is considered a two-fold optimization problem of finding pertinent positives and negatives. Pawelczyk et al make use of an autoencoder architecture for a pretrained classifier performing counterfactual search in the nearest neighbor style [151]. Model-agnostic frameworks are largely found to use decision trees as part of the reasoning mechanism instead of explaining their output.…”

Section: ) Explainability Methodsmentioning

confidence: 99%

“…Labaien et al calculate the number of changes to switch from the original to the selected contrastive sample following the dataset constraints [141]. To estimate faithfulness of the generated counterfactuals, Pawelczyk et al suggest calculating the so-called degree of difficulty of a counterfactual suggestion to measure how costly it is to achieve the state of the given suggestion [151]. Aiming to provide realistic counterfactuals, Sharma et al introduce the counterfactual explanation robustness-based score defined as the expected distance between the input instances and their corresponding counterfactuals [154].…”

Section: ) Evaluation Methodsmentioning

confidence: 99%

“…Indeed, contfactuals are particularly suitable for informing the end-user why a given data example is assigned a particular class label. Thus, the outlined classification-oriented frameworks are evaluated on classifiers based on logistic regression [55], [136], [153], [158], decision trees [46], [80], [122], [140], [150], [155], [159], gradient boosted decision trees [147], support vector machines [131], [138], [146], random forests [81], [86], [142]- [144], neural networks [6], [48], [49], [91], [129], [130], [133], [135], [139], [141], [145], [148], [151], or combinations of these [100], [105], [134], [152], [154], [160]. In three studies [67], [128], [137], the classifiers used in the experiments are not specified.…”

Section: ) Ai Problemmentioning

confidence: 99%

“…[81], [105], [136], [147], [148], [151], [154], [158], [160]. They can be extracted from interpretable featurevalue pairs as a result of pruning in the search space [132].…”

Section: ) Output Representationmentioning

confidence: 99%

See 3 more Smart Citations

A Survey of Contrastive and Counterfactual Explanation Generation Methods for Explainable Artificial Intelligence

et al. 2021

View full text Add to dashboard Cite

A number of algorithms in the field of artificial intelligence offer poorly interpretable decisions. To disclose the reasoning behind such algorithms, their output can be explained by means of so-called evidence-based (or factual) explanations. Alternatively, contrastive and counterfactual explanations justify why the output of the algorithms is not any different and how it could be changed, respectively. It is of crucial importance to bridge the gap between theoretical approaches to contrastive and counterfactual explanation and the corresponding computational frameworks. In this work we conduct a systematic literature review which provides readers with a thorough and reproducible analysis of the interdisciplinary research field under study. We first examine theoretical foundations of contrastive and counterfactual accounts of explanation. Then, we report the state-of-the-art computational frameworks for contrastive and counterfactual explanation generation. In addition, we analyze how grounded such frameworks are on the insights from the inspected theoretical approaches. As a result, we highlight a variety of properties of the approaches under study and reveal a number of shortcomings thereof. Moreover, we define a taxonomy regarding both theoretical and practical approaches to contrastive and counterfactual explanation.

show abstract

Section: ) Explainability Methodsmentioning

confidence: 99%

Section: ) Evaluation Methodsmentioning

confidence: 99%

Section: ) Ai Problemmentioning

confidence: 99%

“…[81], [105], [136], [147], [148], [151], [154], [158], [160]. They can be extracted from interpretable featurevalue pairs as a result of pruning in the search space [132].…”

Section: ) Output Representationmentioning

confidence: 99%

See 2 more Smart Citations

A Survey of Contrastive and Counterfactual Explanation Generation Methods for Explainable Artificial Intelligence

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The ODE layer in the generator transforms ⊕ , the concatenation of a noisy vector and a condition vector , into another latent vector ′ that will be fed into the generator (See Section 3.3). reasons, many web-oriented researchers focus on various tasks on tabular data [10,12,27,30,32,45,56,59,62,63]. In this work, generating realistic synthetic tabular data is of our utmost interest.…”

Section: ) (C)mentioning

confidence: 99%

OCT-GAN: Neural ODE-based Conditional Tabular GANs

Kim

Jeon

Lee

et al. 2021

Proceedings of the Web Conference 2021

View full text Add to dashboard Cite

Synthesizing tabular data is attracting much attention these days for various purposes. With sophisticate synthetic data, for instance, one can augment its training data. For the past couple of years, tabular data synthesis techniques have been greatly improved. Recent work made progress to address many problems in synthesizing tabular data, such as the imbalanced distribution and multimodality problems. However, the data utility of state-of-the-art methods is not satisfactory yet. In this work, we significantly improve the utility by designing our generator and discriminator based on neural ordinary differential equations (NODEs). After showing that NODEs have theoretically preferred characteristics for generating tabular data, we introduce our designs. The NODE-based discriminator performs a hidden vector evolution trajectory-based classification rather than classifying with a hidden vector at the last layer only. Our generator also adopts an ODE layer at the very beginning of its architecture to transform its initial input vector (i.e., the concatenation of a noisy vector and a condition vector in our case) onto another latent vector space suitable for the generation process. We conduct experiments with 13 datasets, including but not limited to insurance fraud detection, online news article prediction, and so on, and our presented method outperforms other state-of-the-art tabular data synthesis methods in many cases of our classification, regression, and clustering experiments. CCS CONCEPTS• Computing methodologies → Machine learning; Neural networks.

show abstract