Building Interpretable Classifiers with Rules using Bayesian Analysis

By Benjamin Letham, Cynthia Rudin. 2012

Subfields: 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 Safety, Robustness & Interpretability

Citations in the corpus

Interpretable Two-level Boolean Rule Learning for Classification (2015, methodology): “In addition to the symbolic approaches above, Bayesian approaches in Letham et al.(2012) and Wang et al.(2015) apply approximate inference algorithms to produce posterior distributions over decision lists; ho wever, the assignment of prior and likelihood may not always be clear, and certain approxim”
Strongly agree or strongly disagree?: Rating features in Support Vector Machines (2016, neutral): “For instance, the CHADS2 score (Gage et al. 2001, Letham et al. 2012) is widely used in medicine to predict the risk of stroke in patients with atrial brillation based on ve di erent symptoms of the patient, whereas the extended CHA2DS2 VASc score (Lip et al. 2010) includes three additional risk…”
Falling Rule Lists (2014, background): “This helps tremendously with computation, and restricts us to building models with only interpretable building blocks (see also Letham et al. , 2012; Wang et al. , 2014).”
Using Decision Lists to Construct Interpretable and Parsimonious Treatment Regimes (2015, methodology): “To obtain an interpretable and parsimonious treatment regime, we use the concept of decision list, which was developed in the computer science literature for representing flexible but interpretable classifiers (Rivest, 1987; Clark and Niblett, 1989; Marchand and Sokolova, 2005; Letham et al., 2012; ”
The fused lasso penalty for learning interpretable medical scoring systems (2017, background): “A similar idea based on Bayesian learning is considered by [14] where the main motivation is to construct simple rules which are interpretable by human experts and can be used by healthcare providers.”
Interpretable Dynamic Treatment Regimes (2016, background): “Decision rules of this form are a special case of tree-based rules, known as decision lists (Rivest, 1987; Marchand and Sokolova, 2005; Letham et al., 2012; Wang and Rudin, 2015; Zhang et al., 2015), that are immediately interpretable in a domain context as they can be expressed in either flow-chart”
Trends and Trajectories for Explainable, Accountable and Intelligible Systems: An HCI Research Agenda (2018, methodology): “Researchers proposed tailored interfaces that explained underlying context-aware rules [41], or provided textual [110, 111, 113, 178, 179] and visual explanations [177] for these rules.”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “The Bayesian Rule Sets (BRS) method [276, 277] is similar to BRL but it uses different probabilities, with the posterior as a Bernoulli distribution, and the prior a Beta distribution whose parameters can be adjusted by end-users to guide BRS toward more interpretable solutions by specifying the des”
A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “The learning process consists of first finding the rules and then ordering them (Letham et al., 2012).”
A Semiquantitative Group Testing Approach for Learning Interpretable Clinical Prediction Rules (2015, methodology): “Recent machine learning research has attempted to learn clinical prediction rules that generalize accurately from large-scale electronic health record data rather than relying on manual development [4], [5].”
Exact Rule Learning via Boolean Compressed Sensing (2013, background): “Bertsimas et al. (2012); Letham et al. (2012) attempt to learn decision trees using mixed integer programming techniques.”
A Provable Algorithm for Learning Interpretable Scoring Systems (2018, background): “A similar idea, also based on Bayesian learning is considered by [16, 30] where the main motivation is to construct simple rules which are interpretable by human experts and can be used by healthcare providers.”
Accountability in algorithmic decision making (2016, methodology): “17 Other methods are being developed in natural language generation (NLG) to output text that explains why or how a decision was reached.”
Detecting concept drift in data streams using model explanation (2018, background): “, 2007), decision rules (Letham et al., 2012) and decision trees (Garćıa et al.”