Cynthia Rudin

Zakk Heile, Hayden McTavish, Varun Babbar et al.

Standard machine learning pipelines often admit many near-optimal models. These "Rashomon sets" pose a range of challenges and opportunities for uncertainty-aware, robust decision making. They allow users to incorporate domain knowledge and preferences that would otherwise be difficult to specify directly in an objective, and they quantify diversity among valid models for a given training dataset and objective function. However, computation of Rashomon sets, even for simple, interpretable model classes such as sparse decision trees, continues to require immense memory and runtime resources. We present PRAXIS, an algorithm to approximate this Rashomon set with orders of magnitude improvement in runtime and memory usage. We validate that PRAXIS regularly recovers almost all of the full Rashomon set. PRAXIS allows researchers and practitioners to scalably model the Rashomon set for real-world datasets. Code for PRAXIS is available at https://github.com/zakk-h/PRAXIS

Zijie J. Wang, Chudi Zhong, Rui Xin et al. · gatech

Given thousands of equally accurate machine learning (ML) models, how can users choose among them? A recent ML technique enables domain experts and data scientists to generate a complete Rashomon set for sparse decision trees--a huge set of almost-optimal interpretable ML models. To help ML practitioners identify models with desirable properties from this Rashomon set, we develop TimberTrek, the first interactive visualization system that summarizes thousands of sparse decision trees at scale. Two usage scenarios highlight how TimberTrek can empower users to easily explore, compare, and curate models that align with their domain knowledge and values. Our open-source tool runs directly in users' computational notebooks and web browsers, lowering the barrier to creating more responsible ML models. TimberTrek is available at the following public demo link: https://poloclub.github.io/timbertrek.

Jon Donnelly, Srikar Katta, Cynthia Rudin et al.

Quantifying variable importance is essential for answering high-stakes questions in fields like genetics, public policy, and medicine. Current methods generally calculate variable importance for a given model trained on a given dataset. However, for a given dataset, there may be many models that explain the target outcome equally well; without accounting for all possible explanations, different researchers may arrive at many conflicting yet equally valid conclusions given the same data. Additionally, even when accounting for all possible explanations for a given dataset, these insights may not generalize because not all good explanations are stable across reasonable data perturbations. We propose a new variable importance framework that quantifies the importance of a variable across the set of all good models and is stable across the data distribution. Our framework is extremely flexible and can be integrated with most existing model classes and global variable importance metrics. We demonstrate through experiments that our framework recovers variable importance rankings for complex simulation setups where other methods fail. Further, we show that our framework accurately estimates the true importance of a variable for the underlying data distribution. We provide theoretical guarantees on the consistency and finite sample error rates for our estimator. Finally, we demonstrate its utility with a real-world case study exploring which genes are important for predicting HIV load in persons with HIV, highlighting an important gene that has not previously been studied in connection with HIV. Code is available at https://github.com/jdonnelly36/Rashomon_Importance_Distribution.

Rui Xin, Chudi Zhong, Zhi Chen et al.

In any given machine learning problem, there may be many models that could explain the data almost equally well. However, most learning algorithms return only one of these models, leaving practitioners with no practical way to explore alternative models that might have desirable properties beyond what could be expressed within a loss function. The Rashomon set is the set of these all almost-optimal models. Rashomon sets can be extremely complicated, particularly for highly nonlinear function classes that allow complex interaction terms, such as decision trees. We provide the first technique for completely enumerating the Rashomon set for sparse decision trees; in fact, our work provides the first complete enumeration of any Rashomon set for a non-trivial problem with a highly nonlinear discrete function class. This allows the user an unprecedented level of control over model choice among all models that are approximately equally good. We represent the Rashomon set in a specialized data structure that supports efficient querying and sampling. We show three applications of the Rashomon set: 1) it can be used to study variable importance for the set of almost-optimal trees (as opposed to a single tree), 2) the Rashomon set for accuracy enables enumeration of the Rashomon sets for balanced accuracy and F1-score, and 3) the Rashomon set for a full dataset can be used to produce Rashomon sets constructed with only subsets of the data set. Thus, we are able to examine Rashomon sets across problems with a new lens, enabling users to choose models rather than be at the mercy of an algorithm that produces only a single model.

Jiachang Liu, Chudi Zhong, Boxuan Li et al.

Over the last century, risk scores have been the most popular form of predictive model used in healthcare and criminal justice. Risk scores are sparse linear models with integer coefficients; often these models can be memorized or placed on an index card. Typically, risk scores have been created either without data or by rounding logistic regression coefficients, but these methods do not reliably produce high-quality risk scores. Recent work used mathematical programming, which is computationally slow. We introduce an approach for efficiently producing a collection of high-quality risk scores learned from data. Specifically, our approach produces a pool of almost-optimal sparse continuous solutions, each with a different support set, using a beam-search algorithm. Each of these continuous solutions is transformed into a separate risk score through a "star ray" search, where a range of multipliers are considered before rounding the coefficients sequentially to maintain low logistic loss. Our algorithm returns all of these high-quality risk scores for the user to consider. This method completes within minutes and can be valuable in a broad variety of applications.

Chudi Zhong, Zhi Chen, Jiachang Liu et al.

In real applications, interaction between machine learning models and domain experts is critical; however, the classical machine learning paradigm that usually produces only a single model does not facilitate such interaction. Approximating and exploring the Rashomon set, i.e., the set of all near-optimal models, addresses this practical challenge by providing the user with a searchable space containing a diverse set of models from which domain experts can choose. We present algorithms to efficiently and accurately approximate the Rashomon set of sparse, generalized additive models with ellipsoids for fixed support sets and use these ellipsoids to approximate Rashomon sets for many different support sets. The approximated Rashomon set serves as a cornerstone to solve practical challenges such as (1) studying the variable importance for the model class; (2) finding models under user-specified constraints (monotonicity, direct editing); and (3) investigating sudden changes in the shape functions. Experiments demonstrate the fidelity of the approximated Rashomon set and its effectiveness in solving practical challenges.

Haiyang Huang, Zhi Chen, Cynthia Rudin

Visual concept discovery has long been deemed important to improve interpretability of neural networks, because a bank of semantically meaningful concepts would provide us with a starting point for building machine learning models that exhibit intelligible reasoning process. Previous methods have disadvantages: either they rely on labelled support sets that incorporate human biases for objects that are "useful," or they fail to identify multiple concepts that occur within a single image. We reframe the concept discovery task as an unsupervised semantic segmentation problem, and present SegDiscover, a novel framework that discovers semantically meaningful visual concepts from imagery datasets with complex scenes without supervision. Our method contains three important pieces: generating concept primitives from raw images, discovering concepts by clustering in the latent space of a self-supervised pretrained encoder, and concept refinement via neural network smoothing. Experimental results provide evidence that our method can discover multiple concepts within a single image and outperforms state-of-the-art unsupervised methods on complex datasets such as Cityscapes and COCO-Stuff. Our method can be further used as a neural network explanation tool by comparing results obtained by different encoders.

Rui Zhang, Rui Xin, Margo Seltzer et al.

Regression trees are one of the oldest forms of AI models, and their predictions can be made without a calculator, which makes them broadly useful, particularly for high-stakes applications. Within the large literature on regression trees, there has been little effort towards full provable optimization, mainly due to the computational hardness of the problem. This work proposes a dynamic-programming-with-bounds approach to the construction of provably-optimal sparse regression trees. We leverage a novel lower bound based on an optimal solution to the k-Means clustering algorithm in 1-dimension over the set of labels. We are often able to find optimal sparse trees in seconds, even for challenging datasets that involve large numbers of samples and highly-correlated features.

Harsh Parikh, Kentaro Hoffman, Haoqi Sun et al.

Epileptiform activity (EA) is associated with worse outcomes including increased risk of disability and death. However, the effect of EA on the neurologic outcome is confounded by the feedback between treatment with anti-seizure medications (ASM) and EA burden. A randomized clinical trial is challenging due to the sequential nature of EA-ASM feedback, as well as ethical reasons. However, some mechanistic knowledge is available, e.g., how drugs are absorbed. This knowledge together with observational data could provide a more accurate effect estimate using causal inference. We performed a retrospective cross-sectional study with 995 patients with the modified Rankin Scale (mRS) at discharge as the outcome and the EA burden defined as the mean or maximum proportion of time spent with EA in six-hour windows in the first 24 hours of electroencephalography as the exposure. We estimated the change in discharge mRS if everyone in the dataset had experienced a certain EA burden and were untreated. We combined pharmacological modeling with an interpretable matching method to account for confounding and EA-ASM feedback. Our matched groups' quality was validated by the neurologists. Having a maximum EA burden greater than 75% when untreated had a 22% increased chance of a poor outcome (severe disability or death), and mild but long-lasting EA increased the risk of a poor outcome by 14%. The effect sizes were heterogeneous depending on pre-admission profile, e.g., patients with hypoxic-ischemic encephalopathy (HIE) or acquired brain injury (ABI) were more affected. Interventions should put a higher priority on patients with an average EA burden higher than 10%, while treatment should be more conservative when the maximum EA burden is low.

Yishay Mansour, Michal Moshkovitz, Cynthia Rudin

Interpretability is an essential building block for trustworthiness in reinforcement learning systems. However, interpretability might come at the cost of deteriorated performance, leading many researchers to build complex models. Our goal is to analyze the cost of interpretability. We show that in certain cases, one can achieve policy interpretability while maintaining its optimality. We focus on a classical problem from reinforcement learning: mazes with $k$ obstacles in $\mathbb{R}^d$. We prove the existence of a small decision tree with a linear function at each inner node and depth $O(\log k + 2^d)$ that represents an optimal policy. Note that for the interesting case of a constant $d$, we have $O(\log k)$ depth. Thus, in this setting, there is no accuracy-interpretability tradeoff. To prove this result, we use a new "compressing" technique that might be useful in additional settings.

Jiachang Liu, Sam Rosen, Chudi Zhong et al.

We consider an important problem in scientific discovery, namely identifying sparse governing equations for nonlinear dynamical systems. This involves solving sparse ridge regression problems to provable optimality in order to determine which terms drive the underlying dynamics. We propose a fast algorithm, OKRidge, for sparse ridge regression, using a novel lower bound calculation involving, first, a saddle point formulation, and from there, either solving (i) a linear system or (ii) using an ADMM-based approach, where the proximal operators can be efficiently evaluated by solving another linear system and an isotonic regression problem. We also propose a method to warm-start our solver, which leverages a beam search. Experimentally, our methods attain provable optimality with run times that are orders of magnitude faster than those of the existing MIP formulations solved by the commercial solver Gurobi.

Quinn Lanners, Harsh Parikh, Alexander Volfovsky et al.

Our goal is to produce methods for observational causal inference that are auditable, easy to troubleshoot, accurate for treatment effect estimation, and scalable to high-dimensional data. We describe a general framework called Model-to-Match that achieves these goals by (i) learning a distance metric via outcome modeling, (ii) creating matched groups using the distance metric, and (iii) using the matched groups to estimate treatment effects. Model-to-Match uses variable importance measurements to construct a distance metric, making it a flexible framework that can be adapted to various applications. Concentrating on the scalability of the problem in the number of potential confounders, we operationalize the Model-to-Match framework with LASSO. We derive performance guarantees for settings where LASSO outcome modeling consistently identifies all confounders (importantly without requiring the linear model to be correctly specified). We also provide experimental results demonstrating the method's auditability, accuracy, and scalability as well as extensions to more general nonparametric outcome modeling.

Ali Behrouz, Mathias Lecuyer, Cynthia Rudin et al.

Sparse decision trees are one of the most common forms of interpretable models. While recent advances have produced algorithms that fully optimize sparse decision trees for prediction, that work does not address policy design, because the algorithms cannot handle weighted data samples. Specifically, they rely on the discreteness of the loss function, which means that real-valued weights cannot be directly used. For example, none of the existing techniques produce policies that incorporate inverse propensity weighting on individual data points. We present three algorithms for efficient sparse weighted decision tree optimization. The first approach directly optimizes the weighted loss function; however, it tends to be computationally inefficient for large datasets. Our second approach, which scales more efficiently, transforms weights to integer values and uses data duplication to transform the weighted decision tree optimization problem into an unweighted (but larger) counterpart. Our third algorithm, which scales to much larger datasets, uses a randomized procedure that samples each data point with a probability proportional to its weight. We present theoretical bounds on the error of the two fast methods and show experimentally that these methods can be two orders of magnitude faster than the direct optimization of the weighted loss, without losing significant accuracy.

Alina Jade Barnett, Zhicheng Guo, Jin Jing et al.

In intensive care units (ICUs), critically ill patients are monitored with electroencephalograms (EEGs) to prevent serious brain injury. The number of patients who can be monitored is constrained by the availability of trained physicians to read EEGs, and EEG interpretation can be subjective and prone to inter-observer variability. Automated deep learning systems for EEG could reduce human bias and accelerate the diagnostic process. However, black box deep learning models are untrustworthy, difficult to troubleshoot, and lack accountability in real-world applications, leading to a lack of trust and adoption by clinicians. To address these challenges, we propose a novel interpretable deep learning model that not only predicts the presence of harmful brainwave patterns but also provides high-quality case-based explanations of its decisions. Our model performs better than the corresponding black box model, despite being constrained to be interpretable. The learned 2D embedded space provides the first global overview of the structure of ictal-interictal-injury continuum brainwave patterns. The ability to understand how our model arrived at its decisions will not only help clinicians to diagnose and treat harmful brain activities more accurately but also increase their trust and adoption of machine learning models in clinical practice; this could be an integral component of the ICU neurologists' standard workflow.

Cynthia Rudin, Chudi Zhong, Lesia Semenova et al.

The Rashomon Effect, coined by Leo Breiman, describes the phenomenon that there exist many equally good predictive models for the same dataset. This phenomenon happens for many real datasets and when it does, it sparks both magic and consternation, but mostly magic. In light of the Rashomon Effect, this perspective piece proposes reshaping the way we think about machine learning, particularly for tabular data problems in the nondeterministic (noisy) setting. We address how the Rashomon Effect impacts (1) the existence of simple-yet-accurate models, (2) flexibility to address user preferences, such as fairness and monotonicity, without losing performance, (3) uncertainty in predictions, fairness, and explanations, (4) reliable variable importance, (5) algorithm choice, specifically, providing advanced knowledge of which algorithms might be suitable for a given problem, and (6) public policy. We also discuss a theory of when the Rashomon Effect occurs and why. Our goal is to illustrate how the Rashomon Effect can have a massive impact on the use of machine learning for complex problems in society.

Harsh Parikh, Quinn Lanners, Zade Akras et al.

Recent statistical and reinforcement learning methods have significantly advanced patient care strategies. However, these approaches face substantial challenges in high-stakes contexts, including missing data, inherent stochasticity, and the critical requirements for interpretability and patient safety. Our work operationalizes a safe and interpretable framework to identify optimal treatment regimes. This approach involves matching patients with similar medical and pharmacological characteristics, allowing us to construct an optimal policy via interpolation. We perform a comprehensive simulation study to demonstrate the framework's ability to identify optimal policies even in complex settings. Ultimately, we operationalize our approach to study regimes for treating seizures in critically ill patients. Our findings strongly support personalized treatment strategies based on a patient's medical history and pharmacological features. Notably, we identify that reducing medication doses for patients with mild and brief seizure episodes while adopting aggressive treatment for patients in intensive care unit experiencing intense seizures leads to more favorable outcomes.

Zhi Chen, Sarah Tan, Urszula Chajewska et al.

Missing values are a fundamental problem in data science. Many datasets have missing values that must be properly handled because the way missing values are treated can have large impact on the resulting machine learning model. In medical applications, the consequences may affect healthcare decisions. There are many methods in the literature for dealing with missing values, including state-of-the-art methods which often depend on black-box models for imputation. In this work, we show how recent advances in interpretable machine learning provide a new perspective for understanding and tackling the missing value problem. We propose methods based on high-accuracy glass-box Explainable Boosting Machines (EBMs) that can help users (1) gain new insights on missingness mechanisms and better understand the causes of missingness, and (2) detect -- or even alleviate -- potential risks introduced by imputation algorithms. Experiments on real-world medical datasets illustrate the effectiveness of the proposed methods.

Zhicheng Guo, Cheng Ding, Duc H. Do et al.

Atrial fibrillation (AF) is the most common type of cardiac arrhythmia. It is associated with an increased risk of stroke, heart failure, and other cardiovascular complications, but can be clinically silent. Passive AF monitoring with wearables may help reduce adverse clinical outcomes related to AF. Detecting AF in noisy wearable data poses a significant challenge, leading to the emergence of various deep learning techniques. Previous deep learning models learn from a single modality, either electrocardiogram (ECG) or photoplethysmography (PPG) signals. However, deep learning models often struggle to learn generalizable features and rely on features that are more susceptible to corruption from noise, leading to sub-optimal performances in certain scenarios, especially with low-quality signals. Given the increasing availability of ECG and PPG signal pairs from wearables and bedside monitors, we propose a new approach, SiamAF, leveraging a novel Siamese network architecture and joint learning loss function to learn shared information from both ECG and PPG signals. At inference time, the proposed model is able to predict AF from either PPG or ECG and outperforms baseline methods on three external test sets. It learns medically relevant features as a result of our novel architecture design. The proposed model also achieves comparable performance to traditional learning regimes while requiring much fewer training labels, providing a potential approach to reduce future reliance on manual labeling.

Chiyu Ma, Brandon Zhao, Chaofan Chen et al.

We present ProtoConcepts, a method for interpretable image classification combining deep learning and case-based reasoning using prototypical parts. Existing work in prototype-based image classification uses a ``this looks like that'' reasoning process, which dissects a test image by finding prototypical parts and combining evidence from these prototypes to make a final classification. However, all of the existing prototypical part-based image classifiers provide only one-to-one comparisons, where a single training image patch serves as a prototype to compare with a part of our test image. With these single-image comparisons, it can often be difficult to identify the underlying concept being compared (e.g., ``is it comparing the color or the shape?''). Our proposed method modifies the architecture of prototype-based networks to instead learn prototypical concepts which are visualized using multiple image patches. Having multiple visualizations of the same prototype allows us to more easily identify the concept captured by that prototype (e.g., ``the test image and the related training patches are all the same shade of blue''), and allows our model to create richer, more interpretable visual explanations. Our experiments show that our ``this looks like those'' reasoning process can be applied as a modification to a wide range of existing prototypical image classification networks while achieving comparable accuracy on benchmark datasets.

Lesia Semenova, Harry Chen, Ronald Parr et al.

The Rashomon set is the set of models that perform approximately equally well on a given dataset, and the Rashomon ratio is the fraction of all models in a given hypothesis space that are in the Rashomon set. Rashomon ratios are often large for tabular datasets in criminal justice, healthcare, lending, education, and in other areas, which has practical implications about whether simpler models can attain the same level of accuracy as more complex models. An open question is why Rashomon ratios often tend to be large. In this work, we propose and study a mechanism of the data generation process, coupled with choices usually made by the analyst during the learning process, that determines the size of the Rashomon ratio. Specifically, we demonstrate that noisier datasets lead to larger Rashomon ratios through the way that practitioners train models. Additionally, we introduce a measure called pattern diversity, which captures the average difference in predictions between distinct classification patterns in the Rashomon set, and motivate why it tends to increase with label noise. Our results explain a key aspect of why simpler models often tend to perform as well as black box models on complex, noisier datasets.

Sully F. Chen, Zhicheng Guo, Cheng Ding et al.

Rapid, reliable, and accurate interpretation of medical time-series signals is crucial for high-stakes clinical decision-making. Deep learning methods offered unprecedented performance in medical signal processing but at a cost: they were compute-intensive and lacked interpretability. We propose Sparse Mixture of Learned Kernels (SMoLK), an interpretable architecture for medical time series processing. SMoLK learns a set of lightweight flexible kernels that form a single-layer sparse neural network, providing not only interpretability, but also efficiency, robustness, and generalization to unseen data distributions. We introduce a parameter reduction techniques to reduce the size of SMoLK's networks while maintaining performance. We test SMoLK on two important tasks common to many consumer wearables: photoplethysmography (PPG) artifact detection and atrial fibrillation detection from single-lead electrocardiograms (ECGs). We find that SMoLK matches the performance of models orders of magnitude larger. It is particularly suited for real-time applications using low-power devices, and its interpretability benefits high-stakes situations.

Emanuele Borgonovo, Francesco Cappelli, Xuefei Lu et al.

Feature importance methods using unrestricted permutations are flawed due to extrapolation errors; such errors appear in all non-trivial variable importance approaches. We propose three new approaches: conditional model reliance and Knockoffs with Gaussian transformation, and restricted ALE plot designs. Theoretical and numerical results show our strategies reduce/eliminate extrapolation.

Pranay Jain, Cheng Ding, Cynthia Rudin et al.

Smart watches and other wearable devices are equipped with photoplethysmography (PPG) sensors for monitoring heart rate and other aspects of cardiovascular health. However, PPG signals collected from such devices are susceptible to corruption from noise and motion artifacts, which cause errors in heart rate estimation. Typical denoising approaches filter or reconstruct the signal in ways that eliminate much of the morphological information, even from the clean parts of the signal that would be useful to preserve. In this work, we develop an algorithm for denoising PPG signals that reconstructs the corrupted parts of the signal, while preserving the clean parts of the PPG signal. Our novel framework relies on self-supervised training, where we leverage a large database of clean PPG signals to train a denoising autoencoder. As we show, our reconstructed signals provide better estimates of heart rate from PPG signals than the leading heart rate estimation methods. Further experiments show significant improvement in Heart Rate Variability (HRV) estimation from PPG signals using our algorithm. We conclude that our algorithm denoises PPG signals in a way that can improve downstream analysis of many different health metrics from wearable devices.

Harsh Parikh, Marco Morucci, Vittorio Orlandi et al.

Experimental and observational studies often lack validity due to untestable assumptions. We propose a double machine learning approach to combine experimental and observational studies, allowing practitioners to test for assumption violations and estimate treatment effects consistently. Our framework proposes a falsification test for external validity and ignorability under milder assumptions. We provide consistent treatment effect estimators even when one of the assumptions is violated. However, our no-free-lunch theorem highlights the necessity of accurately identifying the violated assumption for consistent treatment effect estimation. Through comparative analyses, we show our framework's superiority over existing data fusion methods. The practical utility of our approach is further exemplified by three real-world case studies, underscoring its potential for widespread application in empirical research.

Chloe Qinyu Zhu, Muhang Tian, Lesia Semenova et al.

Prediction of mortality in intensive care unit (ICU) patients typically relies on black box models (that are unacceptable for use in hospitals) or hand-tuned interpretable models (that might lead to the loss in performance). We aim to bridge the gap between these two categories by building on modern interpretable ML techniques to design interpretable mortality risk scores that are as accurate as black boxes. We developed a new algorithm, GroupFasterRisk, which has several important benefits: it uses both hard and soft direct sparsity regularization, it incorporates group sparsity to allow more cohesive models, it allows for monotonicity constraint to include domain knowledge, and it produces many equally-good models, which allows domain experts to choose among them. For evaluation, we leveraged the largest existing public ICU monitoring datasets (MIMIC III and eICU). Models produced by GroupFasterRisk outperformed OASIS and SAPS II scores and performed similarly to APACHE IV/IVa while using at most a third of the parameters. For patients with sepsis/septicemia, acute myocardial infarction, heart failure, and acute kidney failure, GroupFasterRisk models outperformed OASIS and SOFA. Finally, different mortality prediction ML approaches performed better based on variables selected by GroupFasterRisk as compared to OASIS variables. GroupFasterRisk's models performed better than risk scores currently used in hospitals, and on par with black box ML models, while being orders of magnitude sparser. Because GroupFasterRisk produces a variety of risk scores, it allows design flexibility - the key enabler of practical model creation. GroupFasterRisk is a fast, accessible, and flexible procedure that allows learning a diverse set of sparse risk scores for mortality prediction.

Mary V. Bastawrous, Zhi Chen, Alexander C. Ogren et al.

Manipulating the dispersive characteristics of vibrational waves is beneficial for many applications, e.g., high-precision instruments. architected hierarchical phononic materials have sparked promise tunability of elastodynamic waves and vibrations over multiple frequency ranges. In this article, hierarchical unit-cells are obtained, where features at each length scale result in a band gap within a targeted frequency range. Our novel approach, the ``hierarchical unit-cell template method,'' is an interpretable machine-learning approach that uncovers global unit-cell shape/topology patterns corresponding to predefined band-gap objectives. A scale-separation effect is observed where the coarse-scale band-gap objective is mostly unaffected by the fine-scale features despite the closeness of their length scales, thus enabling an efficient hierarchical algorithm. Moreover, the hierarchical patterns revealed are not predefined or self-similar hierarchies as common in current hierarchical phononic materials. Thus, our approach offers a flexible and efficient method for the exploration of new regions in the hierarchical design space, extracting minimal effective patterns for inverse design in applications targeting multiple frequency ranges.

Haiyang Huang, Yingfan Wang, Cynthia Rudin

Parametric dimensionality reduction methods have gained prominence for their ability to generalize to unseen datasets, an advantage that traditional approaches typically lack. Despite their growing popularity, there remains a prevalent misconception among practitioners about the equivalence in performance between parametric and non-parametric methods. Here, we show that these methods are not equivalent -- parametric methods retain global structure but lose significant local details. To explain this, we provide evidence that parameterized approaches lack the ability to repulse negative pairs, and the choice of loss function also has an impact. Addressing these issues, we developed a new parametric method, ParamRepulsor, that incorporates Hard Negative Mining and a loss function that applies a strong repulsive force. This new method achieves state-of-the-art performance on local structure preservation for parametric methods without sacrificing the fidelity of global structural representation. Our code is available at https://github.com/hyhuang00/ParamRepulsor.

Yiyang Sun, Haiyang Huang, Gaurav Rajesh Parikh et al.

Dimension reduction (DR) is inherently non-unique: multiple embeddings can preserve the structure of high-dimensional data equally well while differing in layout or geometry. In this paper, we formally define the Rashomon set for DR -- the collection of `good' embedding -- and show how embracing this multiplicity leads to more powerful and trustworthy representations. Specifically, we pursue three goals. First, we introduce PCA-informed alignment to steer embeddings toward principal components, making axes interpretable without distorting local neighborhoods. Second, we design concept-alignment regularization that aligns an embedding dimension with external knowledge, such as class labels or user-defined concepts. Third, we propose a method to extract common knowledge across the Rashomon set by identifying trustworthy and persistent nearest-neighbor relationships, which we use to construct refined embeddings with improved local structure while preserving global relationships. By moving beyond a single embedding and leveraging the Rashomon set, we provide a flexible framework for building interpretable, robust, and goal-aligned visualizations.

Fakrul Islam Tushar, Ehsan Samei, Cynthia Rudin et al.

Objective: Although medical imaging datasets are increasingly available, abnormal and annotation-intensive findings critical to lung cancer screening, particularly small pulmonary nodules, remain underrepresented and inconsistently curated. Methods: We introduce NodMAISI, an anatomically constrained, nodule-oriented CT synthesis and augmentation framework trained on a unified multi-source cohort (7,042 patients, 8,841 CTs, 14,444 nodules). The framework integrates: (i) a standardized curation and annotation pipeline linking each CT with organ masks and nodule-level annotations, (ii) a ControlNet-conditioned rectified-flow generator built on MAISI-v2's foundational blocks to enforce anatomy- and lesion-consistent synthesis, and (iii) lesion-aware augmentation that perturbs nodule masks (controlled shrinkage) while preserving surrounding anatomy to generate paired CT variants. Results: Across six public test datasets, NodMAISI improved distributional fidelity relative to MAISI-v2 (real-to-synthetic FID range 1.18 to 2.99 vs 1.69 to 5.21). In lesion detectability analysis using a MONAI nodule detector, NodMAISI substantially increased average sensitivity and more closely matched clinical scans (IMD-CT: 0.69 vs 0.39; DLCS24: 0.63 vs 0.20), with the largest gains for sub-centimeter nodules where MAISI-v2 frequently failed to reproduce the conditioned lesion. In downstream nodule-level malignancy classification trained on LUNA25 and externally evaluated on LUNA16, LNDbv4, and DLCS24, NodMAISI augmentation improved AUC by 0.07 to 0.11 at <=20% clinical data and by 0.12 to 0.21 at 10%, consistently narrowing the performance gap under data scarcity.

Simon D. Nguyen, Hayden McTavish, Kentaro Hoffman et al.

Active learning reduces labeling costs by selecting samples that maximize information gain. A dominant framework, Query-by-Committee (QBC), typically relies on perturbation-based diversity by inducing model disagreement through random feature subsetting or data blinding. While this approximates one notion of epistemic uncertainty, it sacrifices direct characterization of the plausible hypothesis space. We propose the complementary approach: Rashomon Ensembled Active Learning (REAL) which constructs a committee by exhaustively enumerating the Rashomon Set of all near-optimal models. To address functional redundancy within this set, we adopt a PAC-Bayesian framework using a Gibbs posterior to weight committee members by their empirical risk. Leveraging recent algorithmic advances, we exactly enumerate this set for the class of sparse decision trees. Across synthetic and established active learning baselines, REAL outperforms randomized ensembles, particularly in moderately noisy environments where it strategically leverages expanded model multiplicity to achieve faster convergence.

Parian Haghighat, Hadis Anahideh, Cynthia Rudin

The existence of multiple, equally accurate models for a given predictive task leads to predictive multiplicity, where a ``Rashomon set'' of models achieve similar accuracy but diverges in their individual predictions. This inconsistency undermines trust in high-stakes applications where we want consistent predictions. We propose three approaches to reduce inconsistency among predictions for the members of the Rashomon set. The first approach is \textbf{outlier correction}. An outlier has a label that none of the good models are capable of predicting correctly. Outliers can cause the Rashomon set to have high variance predictions in a local area, so fixing them can lower variance. Our second approach is local patching. In a local region around a test point, models may disagree with each other because some of them are biased. We can detect and fix such biases using a validation set, which also reduces multiplicity. Our third approach is pairwise reconciliation, where we find pairs of models that disagree on a region around the test point. We modify predictions that disagree, making them less biased. These three approaches can be used together or separately, and they each have distinct advantages. The reconciled predictions can then be distilled into a single interpretable model for real-world deployment. In experiments across multiple datasets, our methods reduce disagreement metrics while maintaining competitive accuracy.

Xiyang Hu, Cynthia Rudin, Margo Seltzer

Decision tree algorithms have been among the most popular algorithms for interpretable (transparent) machine learning since the early 1980's. The problem that has plagued decision tree algorithms since their inception is their lack of optimality, or lack of guarantees of closeness to optimality: decision tree algorithms are often greedy or myopic, and sometimes produce unquestionably suboptimal models. Hardness of decision tree optimization is both a theoretical and practical obstacle, and even careful mathematical programming approaches have not been able to solve these problems efficiently. This work introduces the first practical algorithm for optimal decision trees for binary variables. The algorithm is a co-design of analytical bounds that reduce the search space and modern systems techniques, including data structures and a custom bit-vector library. Our experiments highlight advantages in scalability, speed, and proof of optimality. The code is available at https://github.com/xiyanghu/OSDT.

Stephen Ni-Hahn, Weihan Xu, Jerry Yin et al.

Schenkerian Analysis (SchA) is a uniquely expressive method of music analysis, combining elements of melody, harmony, counterpoint, and form to describe the hierarchical structure supporting a work of music. However, despite its powerful analytical utility and potential to improve music understanding and generation, SchA has rarely been utilized by the computer music community. This is in large part due to the paucity of available high-quality data in a computer-readable format. With a larger corpus of Schenkerian data, it may be possible to infuse machine learning models with a deeper understanding of musical structure, thus leading to more "human" results. To encourage further research in Schenkerian analysis and its potential benefits for music informatics and generation, this paper presents three main contributions: 1) a new and growing dataset of SchAs, the largest in human- and computer-readable formats to date (>140 excerpts), 2) a novel software for visualization and collection of SchA data, and 3) a novel, flexible representation of SchA as a heterogeneous-edge graph data structure.

Chiyu Ma, Jon Donnelly, Wenjun Liu et al.

We present ProtoViT, a method for interpretable image classification combining deep learning and case-based reasoning. This method classifies an image by comparing it to a set of learned prototypes, providing explanations of the form ``this looks like that.'' In our model, a prototype consists of \textit{parts}, which can deform over irregular geometries to create a better comparison between images. Unlike existing models that rely on Convolutional Neural Network (CNN) backbones and spatially rigid prototypes, our model integrates Vision Transformer (ViT) backbones into prototype based models, while offering spatially deformed prototypes that not only accommodate geometric variations of objects but also provide coherent and clear prototypical feature representations with an adaptive number of prototypical parts. Our experiments show that our model can generally achieve higher performance than the existing prototype based models. Our comprehensive analyses ensure that the prototypes are consistent and the interpretations are faithful.

Yingfan Wang, Yiyang Sun, Haiyang Huang et al.

Dimension reduction (DR) algorithms have proven to be extremely useful for gaining insight into large-scale high-dimensional datasets, particularly finding clusters in transcriptomic data. The initial phase of these DR methods often involves converting the original high-dimensional data into a graph. In this graph, each edge represents the similarity or dissimilarity between pairs of data points. However, this graph is frequently suboptimal due to unreliable high-dimensional distances and the limited information extracted from the high-dimensional data. This problem is exacerbated as the dataset size increases. If we reduce the size of the dataset by selecting points for a specific sections of the embeddings, the clusters observed through DR are more separable since the extracted subgraphs are more reliable. In this paper, we introduce LocalMAP, a new dimensionality reduction algorithm that dynamically and locally adjusts the graph to address this challenge. By dynamically extracting subgraphs and updating the graph on-the-fly, LocalMAP is capable of identifying and separating real clusters within the data that other DR methods may overlook or combine. We demonstrate the benefits of LocalMAP through a case study on biological datasets, highlighting its utility in helping users more accurately identify clusters for real-world problems.

Hayden McTavish, Jon Donnelly, Margo Seltzer et al.

Many important datasets contain samples that are missing one or more feature values. Maintaining the interpretability of machine learning models in the presence of such missing data is challenging. Singly or multiply imputing missing values complicates the model's mapping from features to labels. On the other hand, reasoning on indicator variables that represent missingness introduces a potentially large number of additional terms, sacrificing sparsity. We solve these problems with M-GAM, a sparse, generalized, additive modeling approach that incorporates missingness indicators and their interaction terms while maintaining sparsity through l0 regularization. We show that M-GAM provides similar or superior accuracy to prior methods while significantly improving sparsity relative to either imputation or naive inclusion of indicator variables.

Rui Zhang, Rui Xin, Margo Seltzer et al.

Interpretability is crucial for doctors, hospitals, pharmaceutical companies and biotechnology corporations to analyze and make decisions for high stakes problems that involve human health. Tree-based methods have been widely adopted for survival analysis due to their appealing interpretablility and their ability to capture complex relationships. However, most existing methods to produce survival trees rely on heuristic (or greedy) algorithms, which risk producing sub-optimal models. We present a dynamic-programming-with-bounds approach that finds provably-optimal sparse survival tree models, frequently in only a few seconds.

Jon Donnelly, Zhicheng Guo, Alina Jade Barnett et al.

Interpretability is critical for machine learning models in high-stakes settings because it allows users to verify the model's reasoning. In computer vision, prototypical part models (ProtoPNets) have become the dominant model type to meet this need. Users can easily identify flaws in ProtoPNets, but fixing problems in a ProtoPNet requires slow, difficult retraining that is not guaranteed to resolve the issue. This problem is called the "interaction bottleneck." We solve the interaction bottleneck for ProtoPNets by simultaneously finding many equally good ProtoPNets (i.e., a draw from a "Rashomon set"). We show that our framework - called Proto-RSet - quickly produces many accurate, diverse ProtoPNets, allowing users to correct problems in real time while maintaining performance guarantees with respect to the training set. We demonstrate the utility of this method in two settings: 1) removing synthetic bias introduced to a bird identification model and 2) debugging a skin cancer identification model. This tool empowers non-machine-learning experts, such as clinicians or domain experts, to quickly refine and correct machine learning models without repeated retraining by machine learning experts.

Yiyang Sun, Zhi Chen, Vittorio Orlandi et al.

Even if a model is not globally sparse, it is possible for decisions made from that model to be accurately and faithfully described by a small number of features. For instance, an application for a large loan might be denied to someone because they have no credit history, which overwhelms any evidence towards their creditworthiness. In this work, we introduce the Sparse Explanation Value (SEV), a new way of measuring sparsity in machine learning models. In the loan denial example above, the SEV is 1 because only one factor is needed to explain why the loan was denied. SEV is a measure of decision sparsity rather than overall model sparsity, and we are able to show that many machine learning models -- even if they are not sparse -- actually have low decision sparsity, as measured by SEV. SEV is defined using movements over a hypercube, allowing SEV to be defined consistently over various model classes, with movement restrictions reflecting real-world constraints. We proposed the algorithms that reduce SEV without sacrificing accuracy, providing sparse and completely faithful explanations, even without globally sparse models.

Fakrul Islam Tushar, Avivah Wang, Lavsen Dahal et al.

Background: Development of artificial intelligence (AI) models for lung cancer screening requires large, well-annotated low-dose computed tomography (CT) datasets and rigorous performance benchmarks. Purpose: To create a reproducible benchmarking resource leveraging the Duke Lung Cancer Screening (DLCS) and multiple public datasets to develop and evaluate models for nodule detection and classification. Materials & Methods: This retrospective study uses the DLCS dataset (1,613 patients; 2,487 nodules) and external datasets including LUNA16, LUNA25, and NLST-3D. For detection, MONAI RetinaNet models were trained on DLCS (DLCS-De) and LUNA16 (LUNA16-De) and evaluated using the Competition Performance Metric (CPM). For nodule-level classification, we compare five strategies: pretrained models (Models Genesis, Med3D), a self-supervised foundation model (FMCB), and ResNet50 with random initialization versus Strategic Warm-Start (ResNet50-SWS) pretrained with detection-derived candidate patches stratified by confidence. Results: For detection on the DLCS test set, DLCS-De achieved sensitivity 0.82 at 2 false positives/scan (CPM 0.63) versus LUNA16-De (0.62, CPM 0.45). For external validation on NLST-3D, DLCS-De (sensitivity 0.72, CPM 0.58) also outperformed LUNA16-De (sensitivity 0.64, CPM 0.49). For classification across multiple datasets, ResNet50-SWS attained AUCs of 0.71 (DLCS; 95% CI, 0.61-0.81), 0.90 (LUNA16; 0.87-0.93), 0.81 (NLST-3D; 0.79-0.82), and 0.80 (LUNA25; 0.78-0.82), matching or exceeding pretrained/self-supervised baselines. Performance differences reflected dataset label standards. Conclusion: This work establishes a standardized benchmarking resource for lung cancer AI research, supporting model development, validation, and translation. All code, models, and data are publicly released to promote reproducibility.

Elita Lobo, Harvineet Singh, Marek Petrik et al.

Off-policy Evaluation (OPE) methods are a crucial tool for evaluating policies in high-stakes domains such as healthcare, where exploration is often infeasible, unethical, or expensive. However, the extent to which such methods can be trusted under adversarial threats to data quality is largely unexplored. In this work, we make the first attempt at investigating the sensitivity of OPE methods to marginal adversarial perturbations to the data. We design a generic data poisoning attack framework leveraging influence functions from robust statistics to carefully construct perturbations that maximize error in the policy value estimates. We carry out extensive experimentation with multiple healthcare and control datasets. Our results demonstrate that many existing OPE methods are highly prone to generating value estimates with large errors when subject to data poisoning attacks, even for small adversarial perturbations. These findings question the reliability of policy values derived using OPE methods and motivate the need for developing OPE methods that are statistically robust to train-time data poisoning attacks.

Srikar Katta, Harsh Parikh, Cynthia Rudin et al.

Many modern causal questions ask how treatments affect complex outcomes that are measured using wearable devices and sensors. Current analysis approaches require summarizing these data into scalar statistics (e.g., the mean), but these summaries can be misleading. For example, disparate distributions can have the same means, variances, and other statistics. Researchers can overcome the loss of information by instead representing the data as distributions. We develop an interpretable method for distributional data analysis that ensures trustworthy and robust decision-making: Analyzing Distributional Data via Matching After Learning to Stretch (ADD MALTS). We (i) provide analytical guarantees of the correctness of our estimation strategy, (ii) demonstrate via simulation that ADD MALTS outperforms other distributional data analysis methods at estimating treatment effects, and (iii) illustrate ADD MALTS' ability to verify whether there is enough cohesion between treatment and control units within subpopulations to trustworthily estimate treatment effects. We demonstrate ADD MALTS' utility by studying the effectiveness of continuous glucose monitors in mitigating diabetes risks.

Varun Babbar, Hayden McTavish, Cynthia Rudin et al.

Decision tree optimization is fundamental to interpretable machine learning. The most popular approach is to greedily search for the best feature at every decision point, which is fast but provably suboptimal. Recent approaches find the global optimum using branch and bound with dynamic programming, showing substantial improvements in accuracy and sparsity at great cost to scalability. An ideal solution would have the accuracy of an optimal method and the scalability of a greedy method. We introduce a family of algorithms called SPLIT (SParse Lookahead for Interpretable Trees) that moves us significantly forward in achieving this ideal balance. We demonstrate that not all sub-problems need to be solved to optimality to find high quality trees; greediness suffices near the leaves. Since each depth adds an exponential number of possible trees, this change makes our algorithms orders of magnitude faster than existing optimal methods, with negligible loss in performance. We extend this algorithm to allow scalable computation of sets of near-optimal trees (i.e., the Rashomon set).

Hayden McTavish, Zachery Boner, Jon Donnelly et al.

Decision trees are widely used for interpretable machine learning due to their clearly structured reasoning process. However, this structure belies a challenge we refer to as predictive equivalence: a given tree's decision boundary can be represented by many different decision trees. The presence of models with identical decision boundaries but different evaluation processes makes model selection challenging. The models will have different variable importance and behave differently in the presence of missing values, but most optimization procedures will arbitrarily choose one such model to return. We present a boolean logical representation of decision trees that does not exhibit predictive equivalence and is faithful to the underlying decision boundary. We apply our representation to several downstream machine learning tasks. Using our representation, we show that decision trees are surprisingly robust to test-time missingness of feature values; we address predictive equivalence's impact on quantifying variable importance; and we present an algorithm to optimize the cost of reaching predictions.

Quinn Lanners, Cynthia Rudin, Alexander Volfovsky et al.

Data fusion techniques integrate information from heterogeneous data sources to improve learning, generalization, and decision making across data sciences. In causal inference, these methods leverage rich observational data to improve causal effect estimation, while maintaining the trustworthiness of randomized controlled trials. Existing approaches often relax the strong no unobserved confounding assumption by instead assuming exchangeability of counterfactual outcomes across data sources. However, when both assumptions simultaneously fail - a common scenario in practice - current methods cannot identify or estimate causal effects. We address this limitation by proposing a novel partial identification framework that enables researchers to answer key questions such as: Is the causal effect positive or negative? and How severe must assumption violations be to overturn this conclusion? Our approach introduces interpretable sensitivity parameters that quantify assumption violations and derives corresponding causal effect bounds. We develop doubly robust estimators for these bounds and operationalize breakdown frontier analysis to understand how causal conclusions change as assumption violations increase. We apply our framework to the Project STAR study, which investigates the effect of classroom size on students' third-grade standardized test performance. Our analysis reveals that the Project STAR results are robust to simultaneous violations of key assumptions, both on average and across various subgroups of interest. This strengthens confidence in the study's conclusions despite potential unmeasured biases in the data.

Cheng Ding, Zhicheng Guo, Zhaoliang Chen et al.

Foundation models, especially those using transformers as backbones, have gained significant popularity, particularly in language and language-vision tasks. However, large foundation models are typically trained on high-quality data, which poses a significant challenge, given the prevalence of poor-quality real-world data. This challenge is more pronounced for developing foundation models for physiological data; such data are often noisy, incomplete, or inconsistent. The present work aims to provide a toolset for developing foundation models on physiological data. We leverage a large dataset of photoplethysmography (PPG) signals from hospitalized intensive care patients. For this data, we propose SimQuality, a novel self-supervised learning task based on convolutional neural networks (CNNs) as the backbone to enforce representations to be similar for good and poor quality signals that are from similar physiological states. We pre-trained the SimQuality on over 36 million 30-second PPG pairs and then fine-tuned and tested on six downstream tasks using external datasets. The results demonstrate the superiority of the proposed approach on all the downstream tasks, which are extremely important for heart monitoring on wearable devices. Our method indicates that CNNs can be an effective backbone for foundation models that are robust to training data quality.

Ishaan Maitra, Raymond Lin, Eric Chen et al.

Health outcomes depend on complex environmental and sociodemographic factors whose effects change over location and time. Only recently has fine-grained spatial and temporal data become available to study these effects, namely the MEDSAT dataset of English health, environmental, and sociodemographic information. Leveraging this new resource, we use a variety of variable importance techniques to robustly identify the most informative predictors across multiple health outcomes. We then develop an interpretable machine learning framework based on Generalized Additive Models (GAMs) and Multiscale Geographically Weighted Regression (MGWR) to analyze both local and global spatial dependencies of each variable on various health outcomes. Our findings identify NO2 as a global predictor for asthma, hypertension, and anxiety, alongside other outcome-specific predictors related to occupation, marriage, and vegetation. Regional analyses reveal local variations with air pollution and solar radiation, with notable shifts during COVID. This comprehensive approach provides actionable insights for addressing health disparities, and advocates for the integration of interpretable machine learning in public health.

Varun Babbar, Zhicheng Guo, Cynthia Rudin

The performance of machine learning models relies heavily on the quality of input data, yet real-world applications often face significant data-related challenges. A common issue arises when curating training data or deploying models: two datasets from the same domain may exhibit differing distributions. While many techniques exist for detecting such distribution shifts, there is a lack of comprehensive methods to explain these differences in a human-understandable way beyond opaque quantitative metrics. To bridge this gap, we propose a versatile framework of interpretable methods for comparing datasets. Using a variety of case studies, we demonstrate the effectiveness of our approach across diverse data modalities-including tabular data, text data, images, time-series signals -- in both low and high-dimensional settings. These methods complement existing techniques by providing actionable and interpretable insights to better understand and address distribution shifts.

Ruixin Zhang, Jon Donnelly, Zhicheng Guo et al.

Large language models (LLMs) have exhibited remarkable capabilities across various domains. The ability to call external tools further expands their capability to handle real-world tasks. However, LLMs often follow an opaque reasoning process, which limits their usefulness in high-stakes domains where solutions need to be trustworthy to end users. LLMs can choose solutions that are unreliable and difficult to troubleshoot, even if better options are available. We address this issue by forcing LLMs to use external -- more reliable -- tools to solve problems when possible. We present a framework built on the tool-calling capabilities of existing LLMs to enable them to select the most reliable and easy-to-troubleshoot solution path, which may involve multiple sequential tool calls. We refer to this framework as LIT (LLMs with Inspectable Tools). In order to support LIT, we introduce a new and challenging benchmark dataset of 1,300 questions and a customizable set of reliability cost functions associated with a collection of specialized tools. These cost functions summarize how reliable each tool is and how easy it is to troubleshoot. For instance, a calculator is reliable across domains, whereas a linear prediction model is not reliable if there is distribution shift, but it is easy to troubleshoot. A tool that constructs a random forest is neither reliable nor easy to troubleshoot. These tools interact with the Harvard USPTO Patent Dataset and a new dataset of NeurIPS 2023 papers to solve mathematical, coding, and modeling problems of varying difficulty levels. We demonstrate that LLMs can achieve more reliable and informed problem-solving while maintaining task performance using our framework.

Muchang Bahng, Charlie Berens, Jon Donnelly et al.

Species detection is important for monitoring the health of ecosystems and identifying invasive species, serving a crucial role in guiding conservation efforts. Multimodal neural networks have seen increasing use for identifying species to help automate this task, but they have two major drawbacks. First, their black-box nature prevents the interpretability of their decision making process. Second, collecting genetic data is often expensive and requires invasive procedures, often necessitating researchers to capture or kill the target specimen. We address both of these problems by extending prototype networks (ProtoPNets), which are a popular and interpretable alternative to traditional neural networks, to the multimodal, cost-aware setting. We ensemble prototypes from each modality, using an associated weight to determine how much a given prediction relies on each modality. We further introduce methods to identify cases for which we do not need the expensive genetic information to make confident predictions. We demonstrate that our approach can intelligently allocate expensive genetic data for fine-grained distinctions while using abundant image data for clearer visual classifications and achieving comparable accuracy to models that consistently use both modalities.