A new method quantifies feature-map discriminativeness for efficient pruning of deep neural networks.
problem Efficiently pruning deep neural networks to reduce computation while maintaining accuracy.
method Presented a novel mathematical formulation (Discriminant Information, DI) to quantify feature-map discriminativeness, enabling efficient pruning and intra-layer mixed precision quantization.
result Our pruned ResNet50 achieves 44% FLOPs reduction without any Top-1 accuracy loss.
A hierarchy of GNNs based on learnable local features is proposed.
problem Limited understanding of GNN architectures and their systematic construction.
method A hierarchy of GNNs based on aggregation regions is derived, and theoretical results are provided.
result Simple GNN architecture exceeds Weisfeiler-Lehman graph isomorphism test.
New method optimizes kernel feature maps for better classification.
problem High computational and memory complexity of standard kernel methods.
method Discriminant Information criterion for optimizing kernel feature maps.
result Improved optimization and generalization performances over state-of-the-art methods.
Beam search improves feature selection for better model performance.
problem Improving feature selection for better model performance.
method Proposed beam search as a generalization of forward selection for feature selection.
result Beam search can outperform forward selection, especially with correlated features.
Enhances model's ability to distinguish target domain by adding a new class.
problem Improving unsupervised domain adaptation models' discriminative power.
method Training model on data from a new class generated by GAN, repositioning current class data.
result Achieves state-of-the-art performance in various unsupervised domain adaptation scenarios.
We present a growing dimension asymptotic formalism. The perspective in this paper is classification theory and we show that it can accommodate probabilistic networks classifiers, including naive Bayes model and its augmented version. When represented as a Bayesian network these classifiers have an important advantage:…
DSL learns discriminative subgraphs from graphs for robust prediction.
problem Learning discriminative subgraphs from graph data for robust prediction.
method Discriminative Subgraph Learning (DSL) framework that enforces sparsity, connectivity, and high discriminative power.
result DSL improves prediction accuracy by up to 16% compared to baselines.
Measures CNN features' discriminative power for various datasets.
problem Improving feature representation transfer in CNNs.
method Statistical analysis of CNN features across 11 datasets.
result Low and middle level features behave differently under certain conditions.
Develops interpretable low-dimensional kernels with conic discriminant functions.
problem Improving interpretability in kernel-based classification models.
method Gradually constructs simple feature maps leading to interpretable low-dimensional kernels.
result Obtains high accuracy results without extensive hyperparameter tuning.
Method identifies key features for clustering in high-dimensional data.
problem Understanding hidden patterns in high-dimensional data.
method Unsupervised feature selection based on discriminative power.
result 27 key transcription factors identified, 18 known to define cell states.
Paper optimizes graph neural networks for better structural graph classification.
problem Improving graph neural networks for structural graph classification.
method Focus on aggregation functions, specifically sum and histogram-based functions, to enhance discrimination.
result Design of a graph neural network that learns discriminative graph representations.
ControlGAN improves GANs by generating detailed, specific features.
problem Current GANs struggle with generating detailed, specific features.
method ControlGAN separates a feature classifier from a discriminator to generate detailed features.
result ControlGAN generates improved samples with well-controlled features.
New approach to GANs using convex loss functions and kernel-based discriminators.
problem Minimizing the f-divergence between true and fake data distributions.
method Introducing a minimizing general loss viewpoint and using kernel-based discriminators.
result Maximizing the general loss is equivalent to the min-max problem in GAN.
Discriminative clustering learns from both labeled and unlabeled data.
problem Clustering complex datasets with limited labeled data.
method Gradient-based stochastic training and optimal transport with entropic regularization.
result The method can learn feature representations even in fully unsupervised settings.
Proposes CDDA method to adapt models across domains with minimized discrepancy and increased discriminative power.
problem Transfer learning across domains with different distributions.
method CDDA method that minimizes discrepancy and increases discriminative power through latent feature representation.
result Consistently outperforms state-of-the-art methods in cross-domain image classification tasks.
New method improves unsupervised feature learning for natural data.
problem Natural data's correlated and long-tail distribution challenges instance-level contrastive learning.
method Cross-level instance-group discrimination (CLD) to integrate between-instance similarity.
result CLD achieves new state-of-the-art performance on various datasets.
Discriminators can be good feature extractors despite their task focus.
problem Discriminators' features are often considered useless for downstream tasks.
method Theoretical analysis and feature space examination to understand discriminator's role.
result Discriminator features are robust and can prevent mode collapse, making them useful for transfer learning.
Paper introduces new loss functions for Siamese networks using FDA.
problem Training Siamese networks with improved loss functions.
method Proposes Fisher Discriminant Triplet (FDT) and Fisher Discriminant Contrastive (FDC) loss functions based on FDA.
result Shows effectiveness of FDT and FDC on MNIST and histopathology datasets.
Paper proposes a CNN for speech emotion recognition using center loss and reconstruction.
problem Speech emotion recognition (SER) in audio signals.
method Convolutional Neural Network (CNN) with center loss and reconstruction as regularizers.
result Proposed method achieves highly discriminative features for SER.
Random Forests are reinterpreted as generative models to handle missing data and detect outliers.
problem Handling missing features and detecting outliers in Random Forests.
method Interpreting Random Forests as Probabilistic Circuits and applying marginalisation for missing data.
result GeDTs and GeFs can handle missing data and detect outliers under certain assumptions.
New findings show fixed-kernel discriminators are weaker than feature-learning ones.
problem Comparing performance of fixed-kernel and feature-learning discriminators.
method Using function classes F2 and F1, constructing pairs of distributions, and linking IPMs with sliced Wasserstein distances. result Fixed-kernel IPM and SD cannot discriminate certain distributions that feature-learning IPM and SD can.
Proposes joint domain alignment and discriminative feature learning for deep domain adaptation.
problem Reduces domain shift and misclassification of target domain samples.
method Instance-based and center-based discriminative feature learning methods.
result Learning discriminative features in shared feature space significantly boosts deep domain adaptation performance.
Mining discriminative features for graph data has attracted much attention in recent years due to its important role in constructing graph classifiers, generating graph indices, etc. Most measurement of interestingness of discriminative subgraph features are defined on certain graphs, where the structure of graph objec…
Deep learning models benefit from a well-designed feature space to generalize.
problem Improving generalization in deep learning models.
method Feature space design using deep compositional functions, with explicit and implicit regularization.
result Improves generalization performance by mitigating information loss and using learning rate decay as a regularizer.
The paper integrates statistical significance and discriminative power in pattern discovery.
problem Discovering actionable patterns that meet rigorous statistical significance and discriminative power criteria.
method Integrates statistical significance and discriminative power criteria into state-of-the-art algorithms.
result Improves discriminative power and statistical significance of discovered patterns without quality deterioration.
Feature learning forms the cornerstone for tackling challenging learning problems in domains such as speech, computer vision and natural language processing. In this paper, we consider a novel class of matrix and tensor-valued features, which can be pre-trained using unlabeled samples. We present efficient algorithms f…
A new framework for mobile authentication using deep metric learning.
problem Challenges in mobile authentication using behavioral biometrics.
method Deep metric learning, private data protection, flexible training scheduling.
result 95% authentication accuracy on public datasets, robust against attacks.
Feature learning forms the cornerstone for tackling challenging learning problems in domains such as speech, computer vision and natural language processing. In this paper, we consider a novel class of matrix and tensor-valued features, which can be pre-trained using unlabeled samples. We present efficient algorithms f…
New metrics improve quantum ensemble learning efficiency and power.
problem Quantum ensembles' distances poorly understood due to measurement constraints.
method Introduce MMD-k hierarchy of integral probability metrics for quantum ensembles. result MMD-k requires fewer samples for full discriminative power at higher k. Smart Bayes integrates generative and discriminative features for improved classification.
problem Improving classification performance by combining generative and discriminative modeling.
method Integrates generative likelihood-ratio features into a logistic-regression-style classifier.
result Often outperforms logistic regression and Naive Bayes in simulations and real data.
Enhances water disaggregation for parallel appliances using shape features and Bayesian Discriminative Sparse Coding.
problem Accurately discriminate and disaggregate water consumption patterns from parallel appliances.
method Bayesian Discriminative Sparse Coding (BDSC-LP) with Laplace Prior, shape features, Gibbs sampling.
result Extensive experiments validate the effectiveness of the proposed model.
Interventional domain adaptation improves feature transferability by removing spurious correlations.
problem Improper feature transferability due to spurious correlations in domain adaptation.
method Intervention strategy using unlabeled target data to generate counterfactual features and train discriminability invariance.
result Consistent performance improvements over state-of-the-art approaches in various domain adaptation tasks.
Proposes TFDF to learn transferable and discriminative features for unsupervised domain adaptation.
problem Difficult to induce supervised classifier without labeled data in unsupervised domain adaptation.
method TFDF optimizes transferability and discriminability by aligning distributions and minimizing class confusion.
result TFDF achieves better performance on real-world datasets compared to existing methods.
Paper enhances haptic signals distinguishability with boosted technique.
problem Lack of large datasets in haptics domain limits feature extraction.
method General framework for haptic signal analysis, using spectral features and boosted embedding.
result Framework needs less training data and outperforms state-of-the-art.
A new GNN model SPIN achieves state-of-the-art performance on diverse real-world datasets.
problem Graph classification efficiency and accuracy.
method Parallel neighborhood aggregations (PA-GNNs) and SPIN model.
result SPIN model achieves state-of-the-art performance on diverse real-world datasets.
This paper proposes a method to improve few-shot learning by generating multi-level weight-centric features.
problem Improving few-shot learning performance by leveraging both representation power and weight generation capacity.
method A multi-level weight-centric feature learning approach with a weight-centric training strategy and multi-level feature incorporation.
result Significantly outperforms existing methods in low-shot classification benchmarks.
New model selects uncorrelated and discriminative features for unsupervised feature selection.
problem Selecting uncorrelated and discriminative features in high-dimensional data.
method Adaptive graph-based generalized regression model with uncorrelated constraint and ℓ2,1-norm regularization. result The model effectively selects uncorrelated and discriminative features, improving clustering performance.
Improves machine learning models by identifying key variables and smoothing data.
problem Machine learning models often produce sensitive results that lack transparency.
method Data planning procedure to identify discriminating variables and smooth data.
result Demonstrates that this method can improve model sensitivity without sacrificing transparency.
Guided warping augments time series data by aligning features with a teacher.
problem Small time series datasets limit neural network performance.
method Guided warping with a discriminative teacher to augment data deterministically.
result Significant improvement in performance on various time series datasets.
Improves GANs by enforcing diverse feature learning.
problem GANs can collapse to a single configuration and be unstable.
method Enforces diverse feature learning by penalizing correlated features.
result Enforces diverse features, stabilizes training, and improves image synthesis.
A new method for high-dimensional data classification with improved feature selection.
problem High-dimensional data classification with limited interpretability and prediction accuracy.
method Integrates multiclass diagonal discriminant analysis with feature selection.
result Significantly improved prediction accuracy and feature interpretability.
IMKPL learns interpretable prototypes for better classification.
problem Efficient trade-offs between interpretability and prediction accuracy in kernel-based data.
method Local discrimination in feature space, condensed class-homogeneous neighborhoods, combined embedding.
result IMKPL achieves better interpretability and discriminative representation.
We introduce a new discriminant analysis method (Empirical Discriminant Analysis or EDA) for binary classification in machine learning. Given a dataset of feature vectors, this method defines an empirical feature map transforming the training and test data into new data with components having Gaussian empirical distrib…
Fisher's linear discriminant analysis (FLDA) is an important dimension reduction method in statistical pattern recognition. It has been shown that FLDA is asymptotically Bayes optimal under the homoscedastic Gaussian assumption. However, this classical result has the following two major limitations: 1) it holds only fo…
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…
Develops a method to interpret deep learning models by identifying key features.
problem Revealing the decision-making process of blackbox models from raw data to predictions.
method Adversarial attacks to localize discriminative features with statistical guarantees.
result Locally identified features are both biologically plausible and statistically significant.
DFSOS improves sparse discriminant analysis for high-dimensional data.
problem Sparse discriminant analysis in high-dimensional settings with feature selection.
method Deflation-Free Sparse Optimal Scoring (DFSOS) using Bregman iteration and orthogonality-constrained optimization.
result DFSOS achieves comparable or better classification accuracy than deflation-based methods.
Proposes L-Softmax loss for CNNs to improve feature discriminativeness.
problem Lack of explicit feature discriminativeness in cross-entropy loss.
method Introduces L-Softmax loss that encourages intra-class compactness and inter-class separability.
result Deeply learned features with L-Softmax loss are more discriminative, boosting performance.