Fisher loss improves deep domain adaptation by learning discriminative within-class compact and between-class separable representations.
problem Improving deep domain adaptation performance by learning discriminative representations.
method Proposes a Fisher loss to learn discriminative representations that are within-class compact and between-class separable.
result Noticeable improvements in deep domain adaptation performance, e.g., 6.67% absolute improvement in mean accuracy on the Office-Home dataset.
Nonnegative Matrix Factorization (NMF) has been a popular representation method for pattern classification problem. It tries to decompose a nonnegative matrix of data samples as the product of a nonnegative basic matrix and a nonnegative coefficient matrix, and the coefficient matrix is used as the new representation. …
In this paper, we propose a novel linear discriminant analysis criterion via the Bhattacharyya error bound estimation based on a novel L1-norm (L1BLDA) and L2-norm (L2BLDA). Both L1BLDA and L2BLDA maximize the between-class scatters which are measured by the weighted pairwise distances of class means and meanwhile mini…
Probabilistic linear discriminant analysis (PLDA) is a method used for biometric problems like speaker or face recognition that models the variability of the samples using two latent variables, one that depends on the class of the sample and another one that is assumed independent across samples and models the within-c…
Extends L2-norm LDA to 2D inputs using Bhattacharyya bound.
problem L2-norm LDA loses useful image information for 2D inputs.
method 2DBLDA maximizes matrix-based between-class distance and minimizes within-class distance, optimizing Bhattacharyya error bound.
result 2DBLDA improves image recognition and face reconstruction.
Paper optimizes classification of distributions using Wasserstein metric.
problem Classifying instances represented by distributions on a vector space.
method Maximizing Fisher's ratio in the Wasserstein metric space through iterative algorithm.
result The method enhances classification performance and is robust to variations in distribution summaries.
Graph neural networks (GNNs) have become increasingly popular for classification tasks on graph-structured data. Yet, the interplay between graph topology and feature evolution in GNNs is not well understood. In this paper, we focus on node-wise classification, illustrated with community detection on stochastic block m…
Proposes RBLDA for multivariate time series data classification.
problem Classifying multivariate time series data.
method Regularized Bilinear Discriminant Analysis (RBLDA) for multivariate time series data.
result RBLDA achieves the best overall recognition performance on real MTS data sets.
This report works out the details of a closed-form, fully Bayesian, multiclass, openset, generative pattern classifier using multivariate Gaussian likelihoods, with conjugate priors. The generative model has a common within-class covariance, which is proportional to the between-class covariance in the conjugate prior. …
We propose regularization strategies for learning discriminative models that are robust to in-class variations of the input data. We use the Wasserstein-2 geometry to capture semantically meaningful neighborhoods in the space of images, and define a corresponding input-dependent additive noise data augmentation model. …
Sensor drift is a well-known issue in the field of sensors and measurement and has plagued the sensor community for many years. In this paper, we propose a sensor drift correction method to deal with the sensor drift problem. Specifically, we propose a discriminative subspace projection approach for sensor drift reduct…
New hashing method improves document retrieval precision.
problem Efficiently retrieving similar documents from large text databases.
method Pairwise supervised hashing with Bernoulli VAE and unbiased gradient estimator.
result Superior performance compared to existing methods.
Recent advances show that two-dimensional linear discriminant analysis (2DLDA) is a successful matrix based dimensionality reduction method. However, 2DLDA may encounter the singularity issue theoretically and the sensitivity to outliers. In this paper, a generalized Lp-norm 2DLDA framework with regularization for an a…
The "bag-of-frames" approach (BOF), which encodes audio signals as the long-term statistical distribution of short-term spectral features, is commonly regarded as an effective and sufficient way to represent environmental sound recordings (soundscapes) since its introduction in an influential 2007 article. The present …
New framework improves classification accuracy using Pillai's trace and ULDA.
problem Traditional LDA's limitations in noise sensitivity and non-invertible matrices.
method Integrates Pillai's trace with ULDA for a unified classifier.
result Effective control of Type I error rates and improved classification accuracy.
A new method improves few-shot learning by combining ProtoNet with LFD.
problem Few-shot learning struggles with high variance support sets.
method Combines ProtoNet with Local Fisher Discriminant Analysis.
result Superior classification accuracy on miniImageNet and tieredImageNet.
Learning from class-imbalanced data continues to be a common and challenging problem in supervised learning as standard classification algorithms are designed to handle balanced class distributions. While different strategies exist to tackle this problem, methods which generate artificial data to achieve a balanced cla…
D-CBRS manages memory for continual learning by accounting for intra-class diversity.
problem Forgetting in continual learning, especially with class-imbalanced data.
method D-CBRS introduces a novel approach to store instances in memory, considering intra-class diversity.
result D-CBRS outperforms existing methods on data sets with intra-class diversity.
Paper compares credit portfolio risks using robust Bernoulli mixture models.
problem Tackles risk bounds and comparison of credit portfolio losses.
method Uses Bernoulli mixture models with conditional independence and stochastic increasing defaults.
result Provides conditions for comparing conditional default probabilities and portfolio losses.
In this paper, we propose a method for image-set classification based on convex cone models, focusing on the effectiveness of convolutional neural network (CNN) features as inputs. CNN features have non-negative values when using the rectified linear unit as an activation function. This naturally leads us to model a se…
We consider the task of classification in the high dimensional setting where the number of features of the given data is significantly greater than the number of observations. To accomplish this task, we propose a heuristic, called sparse zero-variance discriminant analysis (SZVD), for simultaneously performing linear …
A new measure DCSI quantifies separability for density-based clustering.
problem Quantifying meaningful clusters in data sets.
method Developed a new separability measure DCSI based on separation and connectedness.
result Correctly identifies touching or overlapping classes that do not correspond to meaningful density-based clusters.
This work proposes an adaptive trace lasso regularized L1-norm based graph cut method for dimensionality reduction of Hyperspectral images, called as `Trace Lasso-L1 Graph Cut' (TL-L1GC). The underlying idea of this method is to generate the optimal projection matrix by considering both the sparsity as well as the corr…
Dimensionality reduction (DR) methods have attracted extensive attention to provide discriminative information and reduce the computational burden of the hyperspectral image (HSI) classification. However, the DR methods face many challenges due to limited training samples with high dimensional spectra. To address this …
A new LDA variant improves multi-label classification performance.
problem Improving multi-label classification performance.
method Saliency-based weights redefine between-class and within-class scatter matrices for multi-label classification.
result The proposed method leads to performance improvements in various multi-label classification problems.
Kernel discriminant analysis uses nonlinear embeddings to improve classification.
problem Limited effectiveness of linear discriminant analysis in capturing nonlinear features.
method Study of nonlinear embeddings in kernel discriminant analysis using polynomial and Gaussian kernels, solving generalized eigenvalue problems.
result Polynomial and Gaussian discriminants capture class differences through population moments and randomized projections.
This paper analyzes the landscape of supervised contrastive loss in over-parameterized networks.
problem Understanding the structure of solutions in over-parameterized networks under supervised contrastive loss.
method Analytical approach using unconstrained features model (UFM) to study the solutions of SC loss minimization.
result All local minima of SC loss are global minima in over-parameterized networks, and the minimizer is unique (up to rotation).
Following related work in law and policy, two notions of disparity have come to shape the study of fairness in algorithmic decision-making. Algorithms exhibit treatment disparity if they formally treat members of protected subgroups differently; algorithms exhibit impact disparity when outcomes differ across subgroups,…
New method improves domain generalization by matching object representations.
problem Existing domain generalization methods fail to generalize to unseen domains.
method Proposes matching-based algorithms to match object representations across domains.
result MatchDG algorithm matches ground-truth object representations and improves out-of-domain accuracy.
Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.
problem Understanding and optimizing deep learning training phases.
method Direct measurements on three deepnet architectures across seven datasets.
result Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.
To improve the classification performance in the context of hyperspectral image processing, many works have been developed based on two common strategies, namely the spatial-spectral information integration and the utilization of neural networks. However, both strategies typically require more training data than the cl…
Language models allocate information storage, not collapsing into uniform representations.
problem Incomplete neural collapse in language model representations.
method Analyzing variance and information sharing across 14 models, proving an information floor.
result Within-class variance is allocated information storage, not collapsed into uniform representations.
Neural collapse occurs in normalized features over a Riemannian manifold.
problem Understanding neural collapse in normalized feature models.
method Simplified multi-class classification task to a nonconvex optimization problem over the Riemannian manifold, analyzing the landscape of critical points.
result The only global minimizers are neural collapse solutions, with all other critical points being strict saddles.
This work justifies neural collapse under MSE loss and analyzes the optimization landscape.
problem Understanding neural collapse in deep neural networks under MSE loss.
method Global landscape analysis of vanilla nonconvex MSE loss.
result The only global minimizers are neural collapse solutions.
Effective training of neural networks requires much data. In the low-data regime, parameters are underdetermined, and learnt networks generalise poorly. Data Augmentation alleviates this by using existing data more effectively. However standard data augmentation produces only limited plausible alternative data. Given t…
New method improves transfer and robustness of supervised contrastive learning.
problem Class collapse in supervised contrastive learning leads to poor representation quality.
method Adding a weighted class-conditional InfoNCE loss and a class-conditional autoencoder.
result Improves transfer and robustness on 5 standard datasets and 3 worst-group robustness datasets.
PSC classifier improves HDLSS classification on class-imbalanced data.
problem Classification on high-dimension low-sample-size data with class imbalance.
method Population Structure-learned Classifier (PSC) maximizing inter-class and intra-class scatter matrices.
result PSC outperforms state-of-the-art methods on IHDLSS.
Wide neural networks with weight decay exhibit neural collapse.
problem Proving neural collapse in wide neural networks trained with weight decay.
method Generic guarantees on neural collapse for wide networks with weight decay, proving low training error and balancedness, and bounded conditioning.
result First proof of neural collapse in end-to-end training of wide neural networks with weight decay.
Study reveals significant performance flips in GLOD using repurposed graph classification datasets.
problem Performance discrepancies in graph-level outlier detection using repurposed classification datasets.
method Repurposed binary classification datasets for GLOD; analyzed ROC-AUC performance.
result Performance of GLOD models significantly flips depending on which class is down-sampled.
Generalizes conformal prediction to multiple learnable parameters for efficient prediction sets.
problem Learning valid and efficient prediction sets with low-capacity function classes.
method Constrained empirical risk minimization (ERM) with gradient-based optimization of differentiable surrogate losses and Lagrangians.
result Achieves approximate valid population coverage and near-optimal efficiency within class.
This paper extends neural collapse to class-imbalanced datasets using an unconstrained ReLU feature model.
problem Understanding neural collapse in class-imbalanced datasets with cross-entropy loss.
method Generalized neural collapse to class-imbalanced settings using an unconstrained ReLU feature model.
result Class-means converge to orthogonal vectors with different lengths, and classifier weights align to these vectors.
Proposes a novel classification criterion for high-dimensional data with few samples.
problem Challenges in classifying high-dimensional data with limited samples.
method Tolerance similarity criterion and No-separated Data Maximum Dispersion classifier (NPDMD).
result NPDMD outperforms state-of-the-art methods in various real-world applications.
AGOP mechanism explains deep neural collapse in neural networks.
problem Explaining the rigid structure of data representations in deep neural networks.
method Introducing AGOP and Deep RFM to demonstrate DNC.
result AGOP mechanism causes deep neural collapse in neural networks.
Conditional Random Fields (CRF) are frequently applied for labeling and segmenting sequence data. Morency et al. (2007) introduced hidden state variables in a labeled CRF structure in order to model the latent dynamics within class labels, thus improving the labeling performance. Such a model is known as Latent-Dynamic…
Study shows 'Ordinal Neural Collapse' in deep OR tasks, revealing simple geometric relationships.
problem Understanding neural collapse in deep Ordinal Regression tasks.
method Combining cumulative link models and Unconstrained Feature Model to investigate neural collapse.
result Demonstrates 'Ordinal Neural Collapse' (ONC) with three key properties.
We analyze neural collapse in neural networks, showing that features collapse to vertices of a Simplex ETF.
problem Understanding and optimizing the features learned in the last layer of neural networks during training.
method Simplified unconstrained feature model, studying the global optimization landscape of cross-entropy loss with weight decay.
result The global minimizers of the loss are Simplex ETFs, and other critical points are strict saddles with negative curvature.
This paper explores how kernel methods can explain data effects on neural collapse.
problem Understanding how data affects neural collapse in neural networks.
method Formulating NC1 as a function of kernel, specializing to NNGP and NTK, and exploring a data-aware Gaussian Process kernel.
result The NTK does not represent more collapsed features than the NNGP for Gaussian data, highlighting the limitations of data-independent kernels.
We derive a new radial link for binary classification under shared elliptical distributions.
problem Binary classification under shared-generator elliptical class-conditional distributions.
method We derive the Bayes radial-link family from the within-class radius law and estimate it by a finite fractional-power stochastic-polynomial projection.
result The derived link is asymptotically Bayes-optimal and significantly better than QDA on various benchmarks.