Theoretical study shows adversarial training improves robustness in deep learning models.
problem Ensuring robustness in pre-trained deep learning models.
method Theoretical analysis of adversarial training and feature purification in two-layer neural networks.
result Adversarial training leads to feature purification, making models more robust to attacks.
Adversarial training purifies hidden weights to remove small perturbations.
problem Understanding and removing adversarial perturbations in deep learning models.
method Introducing Feature Purification, a principle that adversarial training aims to remove small dense mixtures in hidden weights.
result Adversarial training can make neural networks robust against small perturbations, even with simple algorithms.
Stochastic defense improves natural classifiers against adversarial attacks.
problem Vulnerability of deep networks to adversarial attacks.
method Long-run MCMC sampling with Energy-Based Model for adversarial purification.
result Balancing memoryless and metastable behavior leads to effective purification and robust classification.
This paper shows diffusion models improve intrusion detection by purifying adversarial examples.
problem Adversarial examples mislead ML intrusion detection systems, leading to false alerts or evasion.
method Used diffusion models to purify adversarial examples in intrusion detection.
result Diffusion models enhance adversarial robustness in intrusion detection without significantly impacting normal performance.
A new law limits kurtosis contrast in balanced mixtures.
problem Kurtosis-based ICA fails in wide, balanced mixtures.
method Proved a redundancy law and showed purification restores contrast.
result Kurtosis contrast obeys O(κmax/Reff) in balanced mixtures. Machine learning models have demonstrated vulnerability to adversarial attacks, more specifically misclassification of adversarial examples. In this paper, we propose a one-off and attack-agnostic Feature Manipulation (FM)-Defense to detect and purify adversarial examples in an interpretable and efficient manner. The i…
The state-of-art DNN structures involve high computation and great demand for memory storage which pose intensive challenge on DNN framework resources. To mitigate the challenges, weight pruning techniques has been studied. However, high accuracy solution for extreme structured pruning that combines different types of …
Models which estimate main effects of individual variables alongside interaction effects have an identifiability challenge: effects can be freely moved between main effects and interaction effects without changing the model prediction. This is a critical problem for interpretability because it permits "contradictory" m…
GNN-CL model improves financial fraud detection using graph neural networks and reinforcement learning.
problem Complex financial fraud detection with high false positive rates.
method Combines GNN, CNN, LSTM, and MLPS to analyze transaction patterns and uses reinforcement learning to filter out noise.
result GNN-CL outperforms existing methods in financial fraud detection, reducing false positives.
The Riemannian Bures metric on the space of (normalized) complex positive matrices is used for parameter estimation of mixed quantum states based on repeated measurements just as the Fisher information in classical statistics. It appears also in the concept of purifications of mixed states in quantum physics. Here we d…
GAMI-Tree uses model-based trees to fit low-order fANOVA models.
problem Fitting interpretable fANOVA models with low-order interactions.
method GAMI-Tree uses model-based trees and a new interaction filtering method.
result GAMI-Tree outperforms EBM and GAMI-Net in predictive performance and interpretability.
A Bayesian framework models adversarial uncertainty for robust machine learning.
problem Vulnerability of machine learning models to adversarial attacks.
method Formal Bayesian framework that models adversarial uncertainty through a stochastic channel, articulating probabilistic assumptions.
result Explicitly modeling adversarial uncertainty leads to improved robustification strategies.
Trimming helps in conformal prediction when it separates anomaly scores.
problem Effectiveness of trimming in conformal prediction under contamination.
method Analyse fixed-threshold trimming as a replacement of the contaminated calibration law with a retained law.
result Trimming helps when it separates anomaly scores, reducing clean-target coverage to a one-dimensional score-CDF transfer problem.
Framework purifies approximate differential privacy to pure differential privacy.
problem Achieving pure differential privacy from approximate differential privacy.
method Randomized post-processing with calibrated noise to eliminate δ parameter.
result First statistically and computationally efficient reduction from approximate DP to pure DP.
EBM life cycle project improves MCMC for image generation, defense, and density modeling.
problem Improving MCMC for diverse EBM applications.
method Three novel MCMC initialization methods for negative samples.
result State-of-the-art performance gains across three applications.
Structured weight pruning is a representative model compression technique of DNNs to reduce the storage and computation requirements and accelerate inference. An automatic hyperparameter determination process is necessary due to the large number of flexible hyperparameters. This work proposes AutoCompress, an automatic…
Machine learning can help us in solving problems in the context big data analysis and classification, as well as in playing complex games such as Go. But can it also be used to find novel protocols and algorithms for applications such as large-scale quantum communication? Here we show that machine learning can be used …
Identifies features most relevant to concept drift in data.
problem Identifying features most relevant to concept drift.
method Distinguishing between drift inducing and faithfully drifting features; deriving minimal subsets of features to characterize drift.
result Derives a detection algorithm for concept drift.
Paper predicts EEG features from acoustic features using RNN and GAN.
problem Predicting EEG features from acoustic features.
method Recurrent Neural Network (RNN) and Generative Adversarial Network (GAN).
result Lower RMSE and normalized RMSE values compared to generating acoustic features from EEG features.
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.
A single pre-trained agent guides feature selection using knockoffs.
problem Feature selection challenges in AI-readiness of data.
method Generates knockoff features and uses reinforcement learning.
result Optimal feature subset identified with reduced dependency on target variable.
Feature networks link ML features via graph structure for enhanced learning.
problem Enhancing feature expressiveness and learning efficiency in machine learning.
method Graph representation of feature vectors, leveraging Fourier and functional analysis.
result Feature networks enable novel, complex feature dependencies.
Approach for selecting features by discarding nuisance and correlated ones.
problem Large datasets with correlated and nuisance features.
method Laplacian score criterion, autoencoder architecture, concrete layer.
result Outperforms similar approaches in clustering performance.
New stability measures for similar features improve feature selection accuracy.
problem Existing stability measures fail to distinguish similar features in highly correlated datasets.
method Introduce new adjusted stability measures that consider feature similarities.
result One new stability measure considers highly similar features as interchangeable.
Counterexamples show HSIC feature selection misses critical features.
problem Feature selection using HSIC misses important features.
method Feature selection via HSIC maximization.
result HSIC feature selection can miss critical features.
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 algorithms select and rank features from MTS without feature extraction.
problem Feature extraction step for MTS classification.
method Directly computes similarity between time series and assesses cluster structure matching labels.
result Techniques match labels well without feature extraction.
Pipeline learns topological features for protein stability prediction.
problem Predicting protein stability using topological features.
method Data-driven method to learn topological features, comparing with expert features.
result Topological features achieve 92%-99% of SME-based models' performance.
In text classification, dictionaries can be used to define human-comprehensible features. We propose an improvement to dictionary features called smoothed dictionary features. These features recognize document contexts instead of n-grams. We describe a principled methodology to solicit dictionary features from a teache…
FeAT improves OOD generalization by learning richer features.
problem Improving feature learning for out-of-distribution (OOD) generalization.
method Feature Augmented Training (FeAT) iteratively augments and retains features from different subsets of training data.
result FeAT effectively learns richer features, boosting OOD performance.
The paper defines and analyzes feature complexity in DNNs, proposing metrics for feature disentanglement and evaluation.
problem Understanding and quantifying the complexity of features learned by deep neural networks.
method Proposes a definition and disentanglement of feature complexity orders, introduces metrics for reliability and over-fitting evaluation.
result Establishes a relationship between feature complexity and DNN performance, and proposes a generic mathematical tool for network compression and knowledge distillation.
Conventional mutual information (MI) based feature selection (FS) methods are unable to handle heterogeneous feature subset selection properly because of data format differences or estimation methods of MI between feature subset and class label. A way to solve this problem is feature transformation (FT). In this study,…
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.
Existing feature selection methods fail to properly account for interactions between features when evaluating feature subsets. In this paper, we attempt to remedy this issue by using orthogonal variance decomposition to evaluate features. The orthogonality of the decomposition allows us to directly calculate the total …
CAN approximates explicit feature interactions for CTR prediction.
problem Learning explicit feature interactions from sparse features.
method Co-Action Network approximates explicit pairwise feature interactions without introducing too many additional parameters.
result CAN outperforms state-of-the-art CTR models and the cartesian product method.
Proposes a new feature selection method integrating feature relationships.
problem Feature selection in machine learning models.
method Integrates feature-feature and feature-target relationships via penalized mRMR.
result Correctly identifies inactive features, reducing false discoveries.
Online feature selection has been an active research area in recent years. We propose a novel diverse online feature selection method based on Determinantal Point Processes (DPP). Our model aims to provide diverse features which can be composed in either a supervised or unsupervised framework. The framework aims to pro…
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.
Feature selection has been proven a powerful preprocessing step for high-dimensional data analysis. However, most state-of-the-art methods tend to overlook the structural correlation information between pairwise samples, which may encapsulate useful information for refining the performance of feature selection. Moreove…
GRANITE unifies feature-based explanation methods to reduce disagreement.
problem Disagreement among feature-based explanation methods.
method GRANITE partitions feature space into regions minimizing interaction and distribution influences.
result Unified and consistent feature explanations.
Orthogonal random features approximate a Bessel kernel, offering sharper bounds than random Fourier features.
problem Approximating Gaussian kernel efficiently for large datasets.
method Use of Haar orthogonal matrices to construct orthogonal random features and analyze their bias and variance.
result Orthogonal random features approximate a Bessel kernel, not the Gaussian kernel, with sharper bounds.
New methods for assessing and visualizing feature groups in machine learning models.
problem Lack of methods for interpreting feature groups in machine learning models.
method Permutation-based, refitting, and Shapley-based techniques for grouped feature importance. Introduced a sequential procedure for identifying stable feature combinations. Developed a combined features effect plot.
result Effective methods for assessing and visualizing the importance and effect of feature groups in machine learning models.
Learning with streaming data has attracted much attention during the past few years. Though most studies consider data stream with fixed features, in real practice the features may be evolvable. For example, features of data gathered by limited-lifespan sensors will change when these sensors are substituted by new ones…
Supporting evidence for adaptive feature program across diverse models.
problem Analyzing feature learning in neural networks.
method Over-parameterized sequence models and feature error measure (FEM).
result FEM is decreasing during training of adaptive feature models.
Hermite polynomials improve private data generation by reducing feature count.
problem Infinite-dimensional features in kernel mean embedding are impractical for private data generation.
method Replace random features with Hermite polynomial features, leveraging their ordered nature.
result Hermite polynomial features yield a more accurate approximation of kernel mean embedding with fewer features.
Feature selection is a technique to screen out less important features. Many existing supervised feature selection algorithms use redundancy and relevancy as the main criteria to select features. However, feature interaction, potentially a key characteristic in real-world problems, has not received much attention. As a…
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.
Learning with feature evolution studies the scenario where the features of the data streams can evolve, i.e., old features vanish and new features emerge. Its goal is to keep the model always performing well even when the features happen to evolve. To tackle this problem, canonical methods assume that the old features …