Introduces RFI for assessing feature importance relative to any subset of features.
problem Lack of nuanced feature importance computation.
method Generalizes PFI and CFI to assess relative feature importance.
result Derives general interpretation rules for RFI.
Interpretability is an important area of research for safe deployment of machine learning systems. One particular type of interpretability method attributes model decisions to input features. Despite active development, quantitative evaluation of feature attribution methods remains difficult due to the lack of ground t…
A framework for quantifying uncertainty in feature importance values.
problem Stable interpretation of feature importance values in machine learning models.
method A novel method based on pairwise comparisons of feature importance values to produce confidence intervals for feature ranks.
result The method produces simultaneous confidence intervals for feature ranks, enabling selection of top-k important features.
Bayesian neural network improves feature selection and prediction.
problem Improving feature selection and prediction accuracy in neural networks.
method BNN-ARD with l2-norm feature importance measure.
result Improves variable selection and predictive performance on real-world data.
Have you ever wondered how your feature space is impacting the prediction of a specific sample in your dataset? In this paper, we introduce Single Sample Feature Importance (SSFI), which is an interpretable feature importance algorithm that allows for the identification of the most important features that contribute to…
New method corrects bias in feature importance measures of GBM.
problem Bias in feature importance measures of GBM.
method Cross-validated unbiased base learners.
result Significant improvement in feature importance measures with minimal computational cost.
BIF assesses feature importance using Dirichlet distribution and Bayesian inference.
problem Quantitative feature importance assessment in statistical models.
method Utilizes Dirichlet distribution for probabilistic feature importance assessment via approximate Bayesian inference.
result Learned importance provides relative significance and confidence quantification of features.
Proposes a framework to assess feature importance without algorithm constraints.
problem Lack of a general framework for assessing feature importance across different algorithms.
method Develops a nonparametric framework for algorithm-agnostic variable importance assessment.
result Valid confidence intervals and testing strategies for variable importance.
This paper shows feature importance remains valid even in low-performing models.
problem Feature importance validity in low-performing machine learning models for biomedical data.
method Experiments with synthetic and real biomedical datasets to compare feature rank stability under different data reductions.
result Feature importance can be maintained even at low performance levels if data size is adequate.
Automates feature selection and weighting in molecular systems.
problem Optimal feature selection and alignment in molecular systems.
method Differentiable Information Imbalance (DII) method for automated feature ranking and scaling.
result Automated feature selection and scaling that preserves information content and interpretability.
Paper compares feature selection methods using GCM and LOCO, showing GCM methods generally outperform LOCO.
problem Feature selection and importance estimation in model-agnostic settings.
method Comparison of feature selection methods related to GCM and LOCO under three model settings.
result GCM-related methods generally outperform LOCO under suitable regularity conditions, as shown by theoretical and empirical results.
Electronic health records are an increasingly important resource for understanding the interactions between patient health, environment, and clinical decisions. In this paper we report an empirical study of predictive modeling of several patient outcomes using three state-of-the-art machine learning methods. Our primar…
The paper benchmarks saliency methods for time series predictions.
problem Benchmarking saliency methods for time series predictions.
method Compared saliency-based interpretability methods across diverse neural architectures.
result Saliency methods fail to reliably identify feature importance over time in time series data.
Redundancy in deep neural network (DNN) models has always been one of their most intriguing and important properties. DNNs have been shown to overparameterize, or extract a lot of redundant features. In this work, we explore the impact of size (both width and depth), activation function, and weight initialization on th…
Introduces a new feature importance measure using Gram-Schmidt decorrelation.
problem Determining feature influence strength and nature in datasets.
method Gram-Schmidt decorrelation and random forest regression.
result Empirical comparison of new estimators with established methods.
Study evaluates feature ranking methods' faithfulness in ML models, improving with dimensionality reduction.
problem Quantifying and improving the faithfulness of feature ranking methods in ML models.
method Evaluation of multiple feature ranking methods, including SHAP, LIME, ALE variance, and LR coefficients, using permutation importance as a baseline.
result Dimensionality reduction improves the faithfulness of feature ranking methods, making permutation importance the most faithful method.
New findings show that minimal feature diversity at neural network initialization is harmful but can be mitigated with noise.
problem The importance of feature diversity at neural network initialization.
method A series of experiments comparing different initialization schemes, including adding noise.
result Minimal feature diversity is harmful but can be mitigated with noise, even standard GPU noise is sufficient.
As machine learning is applied to an increasing variety of complex problems, which are defined by high dimensional and complex data sets, the necessity for task oriented feature learning grows in importance. With the advancement of Deep Learning algorithms, various successful feature learning techniques have evolved. I…
Customer Satisfaction is the most important factors in the industry irrespective of domain. Key Driver Analysis is a common practice in data science to help the business to evaluate the same. Understanding key features, which influence the outcome or dependent feature, is highly important in statistical model building.…
While the success of deep neural networks (DNNs) is well-established across a variety of domains, our ability to explain and interpret these methods is limited. Unlike previously proposed local methods which try to explain particular classification decisions, we focus on global interpretability and ask a universally ap…
In the era of "big data", it is becoming more of a challenge to not only build state-of-the-art predictive models, but also gain an understanding of what's really going on in the data. For example, it is often of interest to know which, if any, of the predictors in a fitted model are relatively influential on the predi…
New method improves model explainability.
problem Improper model explanations fail to reflect true data-generating process.
method Shapley Marginal Surplus for Strong Models
result Significant outperformance in inferential capabilities.
This work explores the importance of model weights and Hessian bias in pruning.
problem Understanding the relative importance of model weights for efficient pruning.
method A principled exploration of pruning, focusing on linear models and neural networks.
result Asymptotic formulas reveal the performance of different pruning methods.
Data augmentation affects feature importance, enhancing learning for neural networks.
problem Understanding the effect of data augmentation on feature importance and learning dynamics.
method Analyzing a two-layer convolutional neural network in a multi-view data model.
result Data augmentation alters feature importance, making certain features more likely to be learned.
OGRe simplifies tensor calculations in general relativity.
problem Complex tensor calculations in general relativity.
method Object-oriented design for tensor calculus, automatic transformations, and optimized algorithms.
result Eliminates user errors and simplifies tensor calculations.
Two new algorithms select matrix rows and columns to preserve distances.
problem Preserving distances in large matrix visualizations.
method Selects rows and columns to preserve distances.
result Preserves distances as closely as possible.
Gaussian OBFS proves strong consistency in feature selection with correlations.
problem Feature selection consistency in the presence of correlations.
method Proves strong consistency of Gaussian OBFS under mild conditions.
result Identifies selected features and rates of convergence for different feature types.
We show that the classification performance of graph convolutional networks (GCNs) is related to the alignment between features, graph, and ground truth, which we quantify using a subspace alignment measure (SAM) corresponding to the Frobenius norm of the matrix of pairwise chordal distances between three subspaces ass…
Current generation of memory-augmented neural networks has limited scalability as they cannot efficiently process data that are too large to fit in the external memory storage. One example of this is lifelong learning scenario where the model receives unlimited length of data stream as an input which contains vast majo…
Magnitude-based features capture interactions between different entities in multispecies spatial data.
problem Capturing interactions between different entities in multispecies spatial data.
method Developing magnitude-based features for multispecies spatial data.
result Identifies distinct neighbourhood types and spatial heterogeneity.
Piecewise Aggregate Approximation (PAA) is a competitive basic dimension reduction method for high-dimensional time series mining. When deployed, however, the limitations are obvious that some important information will be missed, especially the trend. In this paper, we propose two new approaches for time series that u…
This paper compares preprocessing techniques for XGBoost models on various data sets.
problem Improving predictive performance of XGBoost models through optimal data preprocessing.
method Comparison of feature selection, categorical handling, and null imputation methods.
result XGBoost importance by gain is the most consistent and highest-performing method for feature selection.
This work explores how feature decorrelation improves self-supervised learning.
problem Complete and dimensional collapse issues in self-supervised learning.
method Study of a concise framework and connection to feature decorrelation.
result Feature decorrelation improves self-supervised learning.
Derives a biologically plausible neural network for Slow Feature Analysis.
problem Learning latent features from time series data.
method Starting from an SFA objective, derives Bio-SFA with a biologically plausible neural network implementation.
result Validates Bio-SFA on naturalistic stimuli, reproducing interesting properties of brain cells.
RI-based variable ranking and selection outperforms lasso in high-dimensional datasets.
problem Challenges in variable selection and model creation with correlated predictors.
method RI measures for feature ranking and selection, including CRI.Z.
result RI-based methods outperform lasso in high-dimensional datasets, especially with correlated predictors.
The paper develops a theory linking pretraining and fine-tuning in neural networks.
problem Understanding how initialization choices impact feature learning and generalization in neural networks.
method Analytical theory of diagonal linear networks, deriving generalization error as a function of initialization parameters and task statistics.
result Different initialization choices place networks into four fine-tuning regimes with varying abilities to support feature learning and generalization.
Enhances LDL by integrating distance and directional information for more robust label feature representation.
problem Lack of robust label feature representation in LDL tasks, especially with label ambiguity.
method Introduces Structural Anchor Points (SAPs) to capture inter-cluster interactions and a novel LSFs construction strategy, LIFT-SAP.
result Improves LDL performance by 15% on average across 15 real-world datasets.
Combining feature importance estimates improves reliability of machine learning predictions.
problem Lack of consensus on feature importance quantification makes explanations unreliable.
method Proposes a feature importance fusion framework combining multiple quantifiers.
result Feature importance ensembles reduce prediction error by 15%.
A measure of relative importance of variables is often desired by researchers when the explanatory aspects of econometric methods are of interest. To this end, the author briefly reviews the limitations of conventional econometrics in constructing a reliable measure of variable importance. The author highlights the rel…
A new method for measuring conditional feature importance using generative models.
problem Challenges in evaluating feature importance given other feature values.
method Adversarial Random Forest (ARF) for generating on-manifold data points.
result cARFi method yields robust importance scores adaptable for various feature importance notions.
New method disentangles feature importance scores in machine learning.
problem Misinterpretation of feature importance scores due to interactions and dependencies.
method Derive DIP (Disentangled Importance) decomposition of feature importance scores.
result DIP decomposition uniquely separates standalone contributions from interactions and dependencies.
Paper defines feature impact and importance from data, not models.
problem Misinterpretation of feature importance as impact leads to flawed insights.
method Mathematical definitions of feature impact and importance derived from partial dependence curves.
result Feature rankings by these definitions are competitive with existing techniques.
RNNs are vulnerable to adversarial attacks, especially in network traffic.
problem Adversarial attacks on RNNs for IDSs in network traffic.
method Developed new explainability techniques and ARS for comparing IDSs.
result RNNs are vulnerable to adversarial attacks, even in sequential data.
In recent years, a large amount of model-agnostic methods to improve the transparency, trustability and interpretability of machine learning models have been developed. We introduce local feature importance as a local version of a recent model-agnostic global feature importance method. Based on local feature importance…
RAMPART ranks top-k features more accurately than existing methods.
problem Accurate ranking of important features in machine learning.
method Adaptive sequential halving strategy combined with ensembling techniques.
result RAMPART achieves the correct top-k ranking with high probability.
Automatic speaker verification, like every other biometric system, is vulnerable to spoofing attacks. Using only a few minutes of recorded voice of a genuine client of a speaker verification system, attackers can develop a variety of spoofing attacks that might trick such systems. Detecting these attacks using the audi…
Machine learning classifies gravitational wave signals to test General Relativity.
problem Testing General Relativity with gravitational wave signals from binary black hole mergers.
method Convolutional Neural Networks (CNNs) trained on whitened waveforms and response function type observables.
result CNNs improve classification sensitivity by a factor of approximately 33 compared to whitened waveforms.
Improves causal inference with observational data by balancing features and weights.
problem Achieving balance in predictive features for causal inference with observational data.
method Integrates balancing weights into representation learning for causal learning.
result Developed an algorithm for accurate estimation of causal effects.