Mind the Gap: A Generative Approach to Interpretable Feature Selection and Extraction

By Been Kim, Julie Arnold Shah, Finale Doshi-Velez. 2015

114 citations recorded by the source corpus.

Subfields: Representation & Self-Supervised Learning, Safety, Robustness & Interpretability

How researchers see this paper

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

In Representation & Self-Supervised Learning

In Safety, Robustness & Interpretability

Citations in the corpus

Explanation in Human-AI Systems: A Literature Meta-Review, Synopsis of Key Ideas and Publications, and Bibliography for Explainable AI (2019, methodology): “The study by Fails and Olsen's (2003; see also Kim et al., 2015) had as its primary measures the amount of time and the number of training iterations it took for the human-machine system to achieve 97.5% agreement with a gold standard non-interactive classifier.”
Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning (2018, methodology): “Doshi-Velez and Kim [25] identify two classes of evaluation for interpretability: (1) the model is useful for a practical (and perhaps simplified) application used as a proxy to test interpretability [50], [62], [49] or (2) the model is learned using a specific hypothesis class already established t”
A Survey of Methods for Explaining Black Box Models (2018, background): “o its label in its neighborhood; no instances should have a prototype with a dierent label in its neighborhood, and there should be as few prototypes as possible. 36 R. Guidotti et al. Kim et al. in [44,45] design the Bayesian Case Model (BCM) comprehensible predictor c l able to learn prototypes by”
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV) (2017, methodology): “This natural, expected property may be beneﬁcially used to make ﬁne-grained distinctions since relative comparisons between related concepts are a good interpretative tool (Kim et al., 2015; Doshi-Velez et al., 2015; Tibshirani, 1994; Salvatore et al., 2014).”
Interpretable Decision Sets: A Joint Framework for Description and Prediction (2016, background): “For example, prototype vector machines [4], regression models [36, 35], generalized additive models [16, 40, 26, 27], Bayesian case models [19] mind-the-gap models [20], and scoring systems [37] have all been proposed as interpretable models.”
How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation (2018, methodology): “Common types of explanation include regressions with simple, human-simulatable functions [Caruana et al., 2015, Kim et al., 2015a, R¨uping, 2006, Bucilu et al., 2006, Ustun and Rudin, 2016, Doshi-Velez et al., 2015, Kim et al., 2015b, Krakovna and Doshi-Velez, 2016, Hughes et al., 2016, Jung et…”
Simple Rules for Complex Decisions (2017, positive): “Our results add to a growing literature on interpretable machine learning [17, 18, 21, 23, 34].”
Feature-Wise Bias Amplification (2018, methodology): “For example, Kim et al. (2015) use feature selection for interpretability during data exploration.”
A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “gression Linear model Efron et al. (2004) Logistic regression Classication Linear model Berkson (1953) KNN Classication & Regression Nearest Neighbor Freitas (2014) MGM Classication Clustering Kim et al. (2015) AOT Classication Tree-based Si and Zhu (2013) ELSE IF petal length : −inf to 2.45 THE”
Interacting with Predictions: Visual Inspection of Black-box Machine Learning Models (2016, background): “[22, 20] introduce graphical models that allow for interpretability of its features.”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “MGM clustered the data of an input dataset according to the most relevant features.”
Variable Selection in Heterogeneous Datasets: A Truncated-rank Sparse Linear Mixed Model with Applications to Genome-wide Association Studies (2017, background): “In applications, variable selection is broadly used to extract variables that are interpretable or potentially causal [9], [10], especially in biology [11] and medicine [12].”
Rationalizing Neural Predictions (2016, background): “Developing sparse interpretable models holds considerably interest in the broader research community(Letham et al., 2015; Kim et al., 2015).”
Explanation and Justification in Machine Learning : A Survey Or (2017, methodology): “Examples include Supersparse Linear Integer Models [ Ustun and Rudin, 2016 ] and Mind-the-Gap Model [ Kim et al. , 2015 ] .”
Making Tree Ensembles Interpretable: A Bayesian Model Selection Approach (2016, methodology): “For instance, we may be able to design a better prediction model by fixing the bug [14] in the model [15], or we can make a better decision based on the insights on the model [16, 13].”
Towards A Rigorous Science of Interpretable Machine Learning (2017, background): “…interpretability in the context of an application: if the system is useful in either a practical application or a simpliﬁed version of it, then it must be somehow interpretable (e.g. Ribeiro et al. [2016], Lei et al. [2016], Kim et al. [2015a], Doshi-Velez et al. [2015], Kim et al. [2015b]).”