Bayesian principles improve neural additive models for better feature selection and uncertainty.
problem Lack of calibrated uncertainties and feature selection in neural additive models.
method Augmenting NAMs with Bayesian principles to provide credible intervals, feature selection, and interaction ranking.
result Improved performance on tabular datasets and real-world medical tasks.
SHAP explains boosted trees with additively modeled features.
problem Explaining predictions of boosted trees models with additively modeled features.
method SHAP values for additively modeled features in boosted trees models.
result SHAP dependence plot matches partial dependence plot for additively modeled features.
Collaborative Trees model analyzes feature interactions and additive effects.
problem Analyzing complex statistical associations between features and response variables.
method Proposes a novel tree model and its bagging version to decompose mean decrease in impurity and visualize feature contributions.
result Demonstrates the superior capability of the tree model in estimating additive effects and interaction effects.
HARFE approximates sparse additive functions using random features and ridge regression.
problem Approximating high-dimensional sparse additive functions.
method Hard-ridge random feature expansion with sparse ridge regression and hard-thresholding pursuit.
result HARFE method converges with a given error bound and achieves lower error than other algorithms.
FAST optimizes additive segmentation for faster, more interpretable models.
problem Efficiently segmenting and interpreting complex datasets.
method Optimization framework for fast piecewise constant shape functions.
result 2 orders of magnitude faster than state-of-the-art methods.
A new distributed algorithm for fitting sparse additive models with feature division and decorrelation.
problem Fitting high-dimensional sparse additive models efficiently and accurately.
method Divide, decorrelate, and conquer approach.
result Effective and efficient recovery of sparsity patterns and statistical inference for each component.
The generalized partially linear additive model (GPLAM) is a flexible and interpretable approach to building predictive models. It combines features in an additive manner, allowing each to have either a linear or nonlinear effect on the response. However, the choice of which features to treat as linear or nonlinear is …
dnamite simplifies NAMs for feature selection and survival analysis.
problem Handling complex machine learning tasks on large-scale data.
method Python package implementing Neural Additive Models (NAMs) for feature selection and survival analysis.
result dnamite provides a scikit-learn style interface for training NAMs.
GADGET framework decomposes global feature effects using recursive partitioning.
problem Misleading global feature effects when feature interactions are present.
method Generalized additive decomposition of global effects (GADGET) based on recursive partitioning.
result Minimizes interaction-related heterogeneity of local feature effects.
New models improve machine learning accuracy and transparency in finance.
problem Black-box machine learning models lack interpretability in regulated industries.
method Introducing generalized groves of neural additive models with clear feature categories and interactions.
result Generalized groves of neural additive models achieve high accuracy with predominantly linear and sparse nonlinear components.
Regularizes GAMs to improve interpretability by reducing concurvity.
problem Susceptibility of GAMs to concurvity reduces interpretability.
method Proposes a regularizer to penalize pairwise correlations of non-linearly transformed features.
result Improves interpretability and reduces concurvity without sacrificing prediction quality.
RAMs improve GAMs' accuracy by fitting components to subregions of feature space.
problem Subpar accuracy in GAMs due to inability to capture feature interactions.
method Identify subregions of feature space where interactions are minimized, fitting one component per subregion.
result RAMs offer improved expressiveness compared to GAMs while maintaining interpretability.
ProtoNAM models tabular data with neural networks, making predictions transparent.
problem Tabular data analysis using neural networks lacks transparency and accuracy compared to tree-based methods.
method ProtoNAM introduces prototypes into neural networks to model tabular data while maintaining explainability.
result ProtoNAM outperforms existing NN-based GAMs and provides insights into learned feature patterns.
SIAN bridges simple models to neural networks by identifying necessary feature combinations.
problem The gap between simple models and powerful neural networks in performance.
method Feature interaction detection and sparse selection algorithm.
result Competitive performance across multiple tabular datasets with optimal tradeoff.
Fast classification for sparse models, even with correlated features.
problem Sparse classification with many correlated features.
method Linear and quadratic surrogate cuts, priority queue, and analytical solution for exponential loss.
result 2 to 5 times faster than previous approaches, interpretable models with comparable accuracy.
The paper uses SHAP for interpreting machine learning models in hospital data.
problem Interpreting machine learning models in healthcare.
method SHAP for feature importance and feature packing techniques.
result SHAP provides better interpretability of machine learning models in healthcare.
SAGE quantifies feature importance in machine learning models.
problem Understanding the role of individual features in complex models.
method Formalizing predictive power through model-based and universal measures, and introducing SAGE for efficient calculation.
result SAGE assigns more accurate feature importance values than other methods.
AEC method improves XAI for models with collinear features.
problem Collinearity issues in machine learning models.
method Divides multivariate models into univariate models to examine feature effects.
result AEC method is more robust and stable against collinearity.
Oblique BART improves tree-based predictions.
problem Axis-aligned decision rules in BART can be suboptimal.
method Developed an oblique version of BART using data-adaptive hyperplane partitions.
result Oblique BART outperformed axis-aligned BART and other tree methods on benchmarks.
SNAM improves NAM's accuracy and feature selection via group sparsity.
problem Improving interpretability and accuracy in deep learning models.
method Employing group sparsity regularization in neural additive models (SNAM).
result SNAM provably converges to zero training loss and achieves exact support recovery.
Proposes fwelnet to improve prediction using feature information.
problem Improving prediction accuracy in supervised learning settings.
method fwelnet uses feature-weighted elastic net to adapt feature penalties.
result fwelnet outperforms lasso in test MSE and feature selection.
We explain how neural networks learn to solve modular addition tasks.
problem How two-layer neural networks learn to solve modular addition tasks.
method Formalized a diversification condition during training, proving it allows the network to approximate the correct logic for modular addition.
result Neural networks can robustly identify the correct sum through phase symmetry and frequency diversification.
SISR improves feature attribution in complex payoff schemes.
problem Distorted feature attributions due to non-additive payoff functions and high-dimensional feature spaces.
method Sparse Isotonic Shapley Regression (SISR) learns a monotonic transformation to restore additivity and enforces L0 sparsity.
result SISR achieves strong support recovery and stable attributions across various payoff schemes.
Many methods have been proposed for community detection in networks, but most of them do not take into account additional information on the nodes that is often available in practice. In this paper, we propose a new joint community detection criterion that uses both the network edge information and the node features to…
neuralGAM package interprets neural networks by fitting them to feature contributions.
problem difficulty understanding neural network decisions
method Generalized Additive Neural Networks (GAM)
result interpretable Deep Learning model with accurate feature contributions
A method interprets black-box models using an ensemble of gradient boosting machines.
problem Local and global interpretation of black-box models.
method An ensemble of gradient boosting machines (GBMs) to form a generalized additive model.
result Efficiency and properties demonstrated on synthetic and real datasets.
We propose a novel algorithm for greedy forward feature selection for regularized least-squares (RLS) regression and classification, also known as the least-squares support vector machine or ridge regression. The algorithm, which we call greedy RLS, starts from the empty feature set, and on each iteration adds the feat…
Logistic regression (LR) is widely used in clinical prediction because it is simple to deploy and easy to interpret. Nevertheless, being a linear model, LR has limited expressive capability and often has unsatisfactory performance. Generalized additive models (GAMs) extend the linear model with transformations of input…
Unified approach for interpretable regression with flexible modeling.
problem Combining predictive adaptivity with interpretability in heterogeneous data.
method Combining random Fourier features, spectral feature map, principal component analysis, Gaussian mixture model, and cluster-specific generalized additive models.
result Consistently improves upon classical and black-box models across benchmark datasets.
We study the problem of detecting adverse drug events in electronic healthcare records. The challenge in this work is to aggregate heterogeneous data types involving diagnosis codes, drug codes, as well as lab measurements. An earlier framework proposed for the same problem demonstrated promising predictive performance…
AAT separates robust and non-robust features without supervision.
problem Adversarial vulnerability and accuracy reduction in machine learning models.
method Adversarial Asymmetric Training (AAT) algorithm.
result Preserves accuracy and achieves better disentanglement than previous methods.
Diamond method controls FDR for trustworthy feature interaction discovery in ML models.
problem Limited interpretability of ML models due to black box nature.
method Diamond method integrates model-X knockoffs framework to control FDR for non-additive interactions.
result Diamond method ensures accurate discovery of feature interactions with FDR control.
SBAMDT uses adaptive soft splits to model complex decision boundaries.
problem Limited ability of standard decision trees to capture complex decision boundaries.
method Probabilistic additive decision tree model with adaptive soft multivariate splits.
result Demonstrated improved predictive performance on synthetic and real datasets.
Enhances machine learning with background knowledge through feature generation.
problem Machine learning struggles with small datasets.
method Generates features from auxiliary datasets not similar to the original task.
result Significant improvement in various learning tasks.
Sparse generalized additive models (GAMs) are an extension of sparse generalized linear models which allow a model's prediction to vary non-linearly with an input variable. This enables the data analyst build more accurate models, especially when the linearity assumption is known to be a poor approximation of reality. …
Proposes a new model to predict polymer properties by integrating various data types.
problem Inaccurate polymer property prediction due to separate modeling of different data types.
method Multi-modal cascade feature transfer using GCN for chemical structure and molecular descriptors.
result Empirically evaluated model shows higher predictive performance than single-feature approaches.
Sparser Random Feature Models via IMP (ShRIMP) efficiently learns sparse models for high-dimensional data.
problem Learning sparse models for high-dimensional data with sparse variable dependencies.
method Iterative Magnitude Pruning applied to Random Feature Models.
result ShRIMP achieves better or competitive test accuracy compared to state-of-the-art methods.
SurvFD and SurvSHAP-IQ provide interpretable survival models by analyzing feature interactions.
problem Non-additivity of hazard and survival functions limits standard additive explanation methods.
method SurvFD decomposes higher-order effects into time-dependent and time-independent components, extending Shapley interactions to time-indexed functions.
result SurvFD and SurvSHAP-IQ offer a new perspective on survival explanations, explicitly characterizing feature interactions.
Paper extracts features from time series to improve forecasting accuracy.
problem Forecasting time series generated by Itô-type processes with unknown coefficients.
method Statistical adjustment of mixture-type models to extract features from time series data.
result Additional statistical features enhance time series prediction accuracy.
Note that a newer expanded version of this paper is now available at: arXiv:1802.03888 It is critical in many applications to understand what features are important for a model, and why individual predictions were made. For tree ensemble methods these questions are usually answered by attributing importance values to i…
Shapley values criticized for feature selection, leading to new insights.
problem Using Shapley values for feature selection is problematic.
method Introduced and critiqued Shapley values as feature selection tools, using counterexamples and simulations.
result Shapley values may not always align with feature selection goals.
Adversarial training has been successfully applied to build robust models at a certain cost. While the robustness of a model increases, the standard classification accuracy declines. This phenomenon is suggested to be an inherent trade-off. We propose a model that employs feature prioritization by a nonlinear attention…
Proposes a deep neural network for early disk drive failure prediction.
problem Early prediction of disk drive failure using multivariate time series sensor data.
method Enriched features derived from sensor data through transformations, combined with ensemble learning and deep neural network architecture.
result Significantly improved classification accuracy in predicting disk drive failure.
Machine Learning algorithms are increasingly being used in recent years due to their flexibility in model fitting and increased predictive performance. However, the complexity of the models makes them hard for the data analyst to interpret the results and explain them without additional tools. This has led to much rese…
Many biological learning systems such as the mushroom body, hippocampus, and cerebellum are built from sparsely connected networks of neurons. For a new understanding of such networks, we study the function spaces induced by sparse random features and characterize what functions may and may not be learned. A network wi…
Traditional convolutional layers extract features from patches of data by applying a non-linearity on an affine function of the input. We propose a model that enhances this feature extraction process for the case of sequential data, by feeding patches of the data into a recurrent neural network and using the outputs or…
Conformal prediction fails under severe feature turnover in COVID-19 supply chain tasks.
problem Dealing with distribution shift in conformal prediction models.
method Using COVID-19 as a natural experiment across 8 supply chain tasks, analyzing SHAP explanations.
result Coverage drops vary widely (0% to 86.7%) and correlate with single-feature dependence.
Improved ranking method for scarce data with feature info.
problem Ranking items with limited comparisons and feature data.
method Modified RankCentrality using diffusion methods for feature info.
result Meaningful rankings even with scarce comparisons.