Interpretable Clustering via Discriminative Rectangle Mixture Model

Chen, Junxiang; Chang, Yale; Hobbs, Brian D.; Castaldi, Peter J.; Cho, Michael; Silverman, Edwin K.; Dy, Jennifer G.

doi:10.1109/icdm.2016.0097

Cited by 20 publications

(15 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most similar to our work, Bertsimas et al formulate the problem of finding an optimal decision tree to perform clustering as a mixed integer optimization problem and construct an approximate solution via coordinate descent (Bertsimas, Orfanoudaki, and Wiberg 2021). In a similar vein, recent work has looked at building rule sets to explain clusters (Chen et al 2016;Chen 2018;Carrizosa et al 2021b;Pelleg and Moore 2001). In contrast to the existing state of the art, our approach is the first to look at a more general function class to explain clusters.…”

Section: Related Workmentioning

confidence: 69%

Interpretable Clustering via Multi-Polytope Machines

Lawless¹,

Kalagnanam²,

Nguyen³

et al. 2021

Preprint

View full text Add to dashboard Cite

Clustering is a popular unsupervised learning tool often used to discover groups within a larger population such as customer segments, or patient subtypes. However, despite its use as a tool for subgroup discovery and description -few stateof-the-art algorithms provide any rationale or description behind the clusters found. We propose a novel approach for interpretable clustering that both clusters data points and constructs polytopes around the discovered clusters to explain them. Our framework allows for additional constraints on the polytopes -including ensuring that the hyperplanes constructing the polytope are axis-parallel or sparse with integer coefficients. We formulate the problem of constructing clusters via polytopes as a Mixed-Integer Non-Linear Program (MINLP). To solve our formulation we propose a two phase approach where we first initialize clusters and polytopes using alternating minimization, and then use coordinate descent to boost clustering performance. We benchmark our approach on a suite of synthetic and real world clustering problems, where our algorithm outperforms state of the art interpretable and non-interpretable clustering algorithms.

show abstract

Section: Related Workmentioning

confidence: 69%

Interpretable Clustering via Multi-Polytope Machines

Lawless¹,

Kalagnanam²,

Nguyen³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…It achieves superior performance when compared to K-Means. In the study of [21], two different methods with decision rules are applied to describe clusters. One defined the clustering model by providing a set of interpretable rules for each cluster, and the other used matrix decision rules with all the features to construct a clustering model.…”

Section: ) Global Perspectivementioning

confidence: 99%

“…This makes the deep learning model itself not a complete black box and the interpretable approaches based on the DL model are unique. [16] local linear model feature CPAR [17] global decision tree/rule feature Trepan [22], [23], [24], REFNE [26] global decision tree/rule √ feature [19], [20], [21] global decision tree/rule feature Anchors [30], PALM [31], [32] local decision tree/rule √ feature MMD-critic [37] global data point √ data point influence function [38] global data point data point SHAP [40], [42] local Shapley value √ feature [48], [49] global KG feature [50], [51], [52] global KG semantic relations RKGE [53], KPRN [54] global KG decision Path [56] local KG semantic relations [57] local KG semantic relations [64] local NN feature CAM [71], Grad-CAM [70], DeepLIFT [72], LRP [73], IBD [74] local NN feature SVCCA [75] global NN neuronal relations ACD [76] global/l ocal NN feature [78], [79], [80] global NN neuronal semantic [81], [82], [83] local NN feature [85], [86], [87] local NN data/feature…”

Section: A Comparison Analysismentioning

confidence: 99%

Review Study of Interpretation Methods for Future Interpretable Machine Learning

Zhou

2020

IEEE Access

View full text Add to dashboard Cite

In recent years, black-box models have developed rapidly because of their high accuracy. Balancing the interpretability and accuracy is increasingly important. The lack of interpretability severely limits the application of the model in academia and industry. Despite the various interpretable machine learning methods, the perspective and meaning of the interpretation are also different. We provide a review of the current interpretable methods and divide them based on the model being applied. We divide them into two categories: interpretable methods with the self-explanatory model and interpretable methods with external co-explanation. And the interpretable methods with external co-explanation are further divided into subbranch methods based on instances, SHAP, knowledge graph, deep learning, and clustering model. The classification aims to help us understand the model characteristics applied in the interpretable method better. This survey makes the researcher find a suitable model to solve interpretability problems more easily. And the comparison experiments contribute to discovering complementary features from different methods. At the same time, we explore the future challenges and trends of interpretable machine learning to promote the development of interpretable machine learning. INDEX TERMS Deep learning, interpretable machine learning, interpretable methods with the selfexplanatory model, interpretable methods with external co-explanation, influential instance, knowledge graph.

show abstract

“…There also exist methods that can provide post-hoc explanations, for example using a consistent set of decision rules, which lead to a prediction (Lakkaraju, Bach, and Leskovec 2016) or a clustering (Kim, Shah, and Doshi-Velez 2015;Chen et al 2016). These rules together characterise the decision boundary for a cluster, whereas we are interested those patterns that characterise the similarities and differences between components.…”

Section: Related Workmentioning

confidence: 99%

Explainable Data Decompositions

Dalleiger

Vreeken

2020

AAAI

View full text Add to dashboard Cite

Our goal is to discover the components of a dataset, characterize why we deem these components, explain how these components are different from each other, as well as identify what properties they share among each other. As is usual, we consider regions in the data to be components if they show significantly different distributions. What is not usual, however, is that we parameterize these distributions with patterns that are informative for one or more components. We do so because these patterns allow us to characterize what is going on in our data as well as explain our decomposition.We define the problem in terms of a regularized maximum likelihood, in which we use the Maximum Entropy principle to model each data component with a set of patterns. As the search space is large and unstructured, we propose the deterministic DISC algorithm to efficiently discover high-quality decompositions via an alternating optimization approach. Empirical evaluation on synthetic and real-world data shows that DISC efficiently discovers meaningful components and accurately characterises these in easily understandable terms.

show abstract

Interpretable Clustering via Discriminative Rectangle Mixture Model

Cited by 20 publications

References 6 publications

Interpretable Clustering via Multi-Polytope Machines

Interpretable Clustering via Multi-Polytope Machines

Review Study of Interpretation Methods for Future Interpretable Machine Learning

Explainable Data Decompositions

Contact Info

Product

Resources

About