A framework infers feature importance with uncertainties for high-dimensional data.
problem Estimating feature importance in high-dimensional data with uncertainty.
method Shapley value based framework, sub-SAGE, bootstrapping.
result Uncertainties in feature importance can be estimated from bootstrapping.
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.
New framework quantifies variable importance across all good models and is stable across data distribution.
problem Conflicting variable importance conclusions from different models trained on the same data.
method Proposes a new variable importance framework that considers all good models and is stable across data distribution.
result Framework accurately estimates true variable importance and recovers rankings for complex setups.
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.
Paper defines hyperparameter importance for efficient tuning.
problem Computational inefficiency in tuning all hyperparameters.
method Defines hyperparameter importance via subsampling procedures.
result Proposed importance consistent with full data under weak conditions.
Hierarchical-CPI improves variable importance measurement for medical data.
problem Limited interpretability of complex medical models.
method Hierarchical-CPI measures conditional variable importance with statistical control, handling correlated data.
result Hierarchical-CPI outperforms existing methods in medical datasets.
UMFI improves feature importance methods by reducing runtime and enhancing performance.
problem Improving feature importance methods to better explain causal and associative relationships in data.
method Introducing UMFI, which uses dependence removal techniques from AI fairness literature.
result UMFI outperforms MCI, especially in complex data scenarios, and reduces runtime from exponential to super-linear.
t-SNE loses important features in data visualization.
problem t-SNE's loss of important features in data visualization.
method Established mathematical framework to understand t-SNE's loss in different scenarios.
result t-SNE loses important features of data in various scenarios.
Importance sampling is often used in machine learning when training and testing data come from different distributions. In this paper we propose a new variant of importance sampling that can reduce the variance of importance sampling-based estimates by orders of magnitude when the supports of the training and testing d…
A new method for variable importance measures without impossible data.
problem Using impossible data for variable importance measures in black box models.
method Cohort Shapley, a method grounded in economic game theory using only observed data.
result Cohort Shapley provides a more trustworthy explanation of black box models' decisions.
New method compares feature importance and rule extraction for text data interpretability.
problem Unexpected differences in explanations from feature importance and rule extraction methods.
method Proposes a new approach to compare explanations from different methods.
result Different methods can lead to unexpected explanations, even for simple models.
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.
The paper analyzes and improves privacy in machine learning through importance sampling.
problem Ensuring privacy in machine learning while maintaining utility and efficiency.
method Individualized privacy analysis of importance sampling, proposing two approaches for constructing sampling distributions.
result Proposed approaches optimize privacy-efficiency trade-off and outperform uniform sampling.
Cluster LOCO: A model-agnostic feature importance score for interpreting cluster outputs
problem Interpreting and auditing cluster outputs
method Cluster LOCO (Leave-One-Covariate-Out)
result More reliably recovers informative features than existing methods
Paper uses Shapley values to explain how clusters of training data affect predictions.
problem Explaining how training data clusters impact predictions from black-box models.
method Extends Shapley values to cluster importance, using coalitional game theory.
result Shows how different clusters of training data contribute to model predictions.
StaPLR improves Alzheimer's disease classification by identifying important MRI scan types and measures.
problem Classifying Alzheimer's disease using multi-source MRI data.
method Stacked penalized logistic regression (StaPLR) with hierarchical multi-view structure and new view importance measure.
result StaPLR identifies the most important MRI scan types and measures for Alzheimer's disease classification.
Enhances functional classifier performance with new tree-based methods and unbiased feature importance assessment.
problem Challenges of high-dimensional functional data and biased feature importance assessment.
method Augmented functional classification trees and random forests with ad-hoc conditional permutations for unbiased feature importance.
result Significant enhancement in predictive power of functional classifiers through new feature importance assessment.
The Importance Weighted Auto Encoder (IWAE) objective has been shown to improve the training of generative models over the standard Variational Auto Encoder (VAE) objective. Here, we derive importance weighted extensions to AVB and AAE. These latent variable models use implicitly defined inference networks whose approx…
SSFI identifies key features affecting a single data point's prediction.
problem Understanding feature importance for a single data point.
method Interpretable algorithm using Random Forest's prediction path.
result Relative ranking of features impacting a data point's prediction.
Optimizes weights for better model performance in shifting data.
problem Improper importance weighting leads to poor model performance in data shifts.
method Interprets weights as a bias-variance trade-off and optimizes them simultaneously with model parameters.
result Optimizing weights significantly improves model generalization performance.
A new method for feature importance inference without data splitting.
problem Feature importance inference for machine learning models.
method Minipatch ensembles for model-agnostic, distribution-free inference.
result Asymptotic validity of confidence intervals without data splitting.
A new method uses nearest neighbors for importance weighting.
problem Data covariate shift problems in machine learning.
method Nearest neighbor classification scheme for determining importance weights.
result Demonstrated effectiveness through comparative experiments on various classification tasks.
Study predicts academic achievement using students' support networks.
problem Predicting academic achievement in college students.
method Decision tree and random forest algorithms applied to Ties data.
result Different types of support are important for different demographics and genders.
CPI overcomes limitations of permutation importance by providing accurate variable selection.
problem Misidentification of unimportant variables in complex models due to covariate correlations.
method Developed a model agnostic and computationally lean Conditional Permutation Importance (CPI) approach.
result CPI provides accurate type-I error control and more parsimonious variable selection.
DEDACT breaks down feature importance into direct and associative components.
problem Lack of clear distinction between direct and associative feature importance.
method DEDACT framework to decompose direct and associative importance measures.
result Provides insight into sources of prediction-relevant information and feature pathways.
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.
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%.
We consider the problem of off-policy evaluation in Markov decision processes. Off-policy evaluation is the task of evaluating the expected return of one policy with data generated by a different, behavior policy. Importance sampling is a technique for off-policy evaluation that re-weights off-policy returns to account…
Proposes VEESA pipeline for interpreting ML models with functional data.
problem Interpreting ML models with functional data inputs for high-consequence applications.
method VEESA pipeline using efPCA and PFI.
result Provides explanation of model's use of functional data variability for prediction.
Feature selection methods are widely used in order to solve the 'curse of dimensionality' problem. Many proposed feature selection frameworks, treat all data points equally; neglecting their different representation power and importance. In this paper, we propose an unsupervised hypergraph feature selection method via …
This paper explores emerging methods for estimating variable importance in machine learning.
problem Estimating the importance of variables in machine learning models.
method Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine (SVM), Predictive Error Function (PERF), Random Forest (RF), Extreme Gradient Boosting (XGBOOST).
result PERF and RF showed the best performance in highly correlated data, while PERF and XGBOOST performed poorly on small data sizes.
Federated Learning is introduced to protect privacy by distributing training data into multiple parties. Each party trains its own model and a meta-model is constructed from the sub models. In this way the details of the data are not disclosed in between each party. In this paper we investigate the model interpretation…
Variable importance is central to scientific studies, including the social sciences and causal inference, healthcare, and other domains. However, current notions of variable importance are often tied to a specific predictive model. This is problematic: what if there were multiple well-performing predictive models, and …
Sharp analysis of out-of-distribution error in overparameterized models with importance weights.
problem Understanding and quantifying the degradation of performance in overparameterized models when faced with underrepresented data.
method Sharp analysis of an overparameterized Gaussian mixture model with spurious features and cost-sensitive interpolating solutions incorporating importance weights.
result Characterization of a novel tradeoff between worst-case robustness and average accuracy as a function of importance weight magnitude.
We improve likelihood-free inference using distillation of importance sampling.
problem Challenging likelihood-free inference with high-dimensional, dependent posterior.
method Approximate posterior with normalizing flows trained on likelihood-free importance sampling.
result Improved accuracy in inference without needing summary statistics.
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.
New method corrects correlation bias in feature importance.
problem Correlations between features in high-dimensional data bias statistical and machine learning models.
method Pairwise Permutation Algorithm (PPA)
result PPA corrects correlation bias, identifying biological relevant biomarkers.
Proposes SOVR loss to improve adversarial robustness by increasing logit margins.
problem Adversarial training's difficulty in robustness against sophisticated attacks.
method Introduces SOVR loss function that switches from cross-entropy to one-vs-the-rest loss for important samples.
result SOVR loss increases logit margins of important samples, improving robustness against Auto-Attack.
Adaptive transfer learning improves time series classification.
problem Improving generalization in time series classification.
method Proposed a general adaptive transfer learning framework for multi-view time series data.
result Framework maintains desirable generalization ability.
CXPlain provides accurate, fast feature importance estimates and uncertainty quantification for machine learning models.
problem Accurate and fast feature importance estimates for high-dimensional data with uncertainty quantification.
method CXPlain models learn to estimate feature importance as a causal learning task, using bootstrap ensembling to quantify uncertainty.
result CXPlain is significantly more accurate and faster than existing methods for estimating feature importance.
Counterfactual approach explains AI decisions using causal data inputs.
problem Explain AI decisions made by data-driven models.
method Define explanations as causal data inputs that drive decisions and are irreducible.
result Counterfactual explanations better communicate decision-making than importance weights.
New method improves scalability of Shapley value without sacrificing utility.
problem Improving scalability of Shapley value for practical applications.
method Approximation of Shapley value using KNN surrogate over pre-trained feature embeddings.
result KNN-based Shapley value approximation achieves comparable utility to existing methods with significant scalability improvements.
A novel approach combines feature importance scores with deep learning for forex price prediction.
problem Improving forex price prediction using deep learning models.
method Feature importance recap combined with stacking models.
result Proper feature selection significantly improves model performance.
TF-MoDISco (Transcription Factor Motif Discovery from Importance Scores) is an algorithm for identifying motifs from basepair-level importance scores computed on genomic sequence data. This technical note focuses on version v0.5.6.5. The implementation is available at https://github.com/kundajelab/tfmodisco/tree/v0.5.6…
A new method assesses algorithmic fairness using game theory.
problem Evaluating algorithmic fairness without proprietary data.
method Cohort Shapley value, a game-theoretic approach.
result Identifies individual impact of protected attributes.
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…
Global Sensitivity Analysis improves feature importance ranking in Random Forests.
problem Improving feature importance ranking in Random Forests.
method Applying Global Sensitivity Analysis to Random Forests for feature ranking.
result Our method provides a novel way to rank features based on their importance.
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.