Unsupervised feature selection (FS) is essential for high-dimensional learning tasks where labels are not available. It helps reduce noise, improve generalization, and enhance interpretability. However, most existing unsupervised FS methods evaluate features in isolation, even though informative signals often emerge from groups of related features. For example, adjacent pixels, functionally connected brain regions, or correlated financial indicators tend to act together, making independent evaluation suboptimal. Although some methods attempt to capture group structure, they typically rely on predefined partitions or label supervision, limiting their applicability. We propose GroupFS, an end-to-end, fully differentiable framework that jointly discovers latent feature groups and selects the most informative groups among them, without relying on fixed a priori groups or label supervision. GroupFS enforces Laplacian smoothness on both feature and sample graphs and applies a group sparsity regularizer to learn a compact, structured representation. Across nine benchmarks spanning images, tabular data, and biological datasets, GroupFS consistently outperforms state-of-the-art unsupervised FS in clustering and selects groups of features that align with meaningful patterns.

Hyperspectral Imaging (HSI) captures rich spectral information across contiguous wavelength bands, supporting applications in precision agriculture, environmental monitoring, and autonomous driving. However, its high dimensionality poses computational challenges, particularly in real-time or resource-constrained settings. While prior band selection methods attempt to reduce complexity, they often rely on separate preprocessing steps and lack alignment with downstream tasks. We propose two novel supervised, embedded methods for task-specific HSI band selection that integrate directly into deep learning models. By embedding band selection within the training pipeline, our methods eliminate the need for separate preprocessing and ensure alignment with the target task. Extensive experiments on three remote sensing benchmarks and an autonomous driving dataset show that our methods achieve state-of-the-art performance while selecting only a minimal number of bands. These results highlight the potential of efficient, task-specific HSI pipelines for practical deployment.

Temporal alignment of sequences is a fundamental challenge in many applications, such as computer vision and bioinformatics, where local time shifting needs to be accounted for. Misalignment can lead to poor model generalization, especially in high-dimensional sequences. Existing methods often struggle with optimization when dealing with high-dimensional sparse data, falling into poor alignments. Feature selection is frequently used to enhance model performance for sparse data. However, a fixed set of selected features would not generally work for dynamically changing sequences and would need to be modified based on the state of the sequence. Therefore, modifying the selected feature based on contextual input would result in better alignment. Our suggested method, Conditional Deep Canonical Temporal Time Warping (CDCTW), is designed for temporal alignment in sparse temporal data to address these challenges. CDCTW enhances alignment accuracy for high dimensional time-dependent views by performing dynamic time warping on data embedded in maximally correlated subspace, which handles sparsity with a novel feature selection method. We validate the effectiveness of CDCTW through extensive experiments on various datasets, demonstrating superior performance over previous techniques.

Combining data from different sources can improve data analysis tasks such as clustering. However, most of the current multi-view clustering methods are limited to specific domains or rely on a suboptimal and computationally intensive two-stage process of representation learning and clustering. We propose an end-to-end deep learning-based multi-view clustering framework for general data types (such as images and tables). Our approach involves generating meaningful fused representations using a novel permutation-based canonical correlation objective. We provide a theoretical analysis showing how the learned embeddings approximate those obtained by supervised linear discriminant analysis (LDA). Cluster assignments are learned by identifying consistent pseudo-labels across multiple views. Additionally, we establish a theoretical bound on the error caused by incorrect pseudo-labels in the unsupervised representations compared to LDA. Extensive experiments on ten multi-view clustering benchmark datasets provide empirical evidence for the effectiveness of the proposed model.

Training and fine-tuning large language models (LLMs) come with challenges related to memory and computational requirements due to the increasing size of the model weights and the optimizer states. To tackle these challenges, various techniques have been developed, such as low-rank adaptation (LoRA), which involves introducing a parallel trainable low-rank matrix to the fixed pre-trained weights at each layer. However, these methods often fall short compared to the full-rank weight training approach, as they restrict the parameter search to a low-rank subspace. This limitation can disrupt training dynamics and may require a full-rank warm start to mitigate the impact. In this paper, we introduce a new method inspired by a phenomenon we formally prove: as training progresses, the rank of the estimated layer gradients gradually decreases and asymptotically approaches rank one. Leveraging this, our approach involves adaptively reducing the rank of the gradients during Adam optimization steps, using an efficient online-updating low-rank projections rule. We further present a randomized-svd scheme for efficiently finding the projection matrix. Our technique enables full-parameter fine-tuning with adaptive low-rank gradient updates, significantly reducing overall memory requirements during training compared to state-of-the-art methods while improving model performance in both pretraining and fine-tuning. Finally, we provide a convergence analysis of our method and demonstrate its merits for training and fine-tuning language and biological foundation models.

Embedding and visualization techniques are essential for analyzing high-dimensional data, but they often struggle with complex data governed by multiple latent variables, potentially distorting key structural characteristics. This paper considers scenarios where the observed features can be partitioned into mutually exclusive subsets, each capturing a different smooth substructure. In such cases, visualizing the data based on each feature partition can better characterize the underlying processes and structures in the data, leading to improved interpretability. To partition the features, we propose solving an optimization problem that promotes graph Laplacian-based smoothness in each partition, thereby prioritizing partitions with simpler geometric structures. Our approach generalizes traditional embedding and visualization techniques, allowing them to learn multiple embeddings simultaneously. We establish that if several independent or partially dependent manifolds are embedded in distinct feature subsets in high-dimensional space, then our framework can reliably identify the correct subsets with theoretical guarantees. Finally, we demonstrate the effectiveness of our approach in extracting multiple low-dimensional structures and partially independent processes from both simulated and real data.

Deep neural networks often under-perform on tabular data due to their sensitivity to irrelevant features and a spectral bias toward smooth, low-frequency functions. These limitations hinder their ability to capture the sharp, high-frequency signals that often define tabular structure, especially under limited labeled samples. While self-supervised learning (SSL) offers promise in such settings, it remains challenging in tabular domains due to the lack of effective data augmentations. We propose a hybrid autoencoder that combines a neural encoder with an oblivious soft decision tree (OSDT) encoder, each guided by its own stochastic gating network that performs sample-specific feature selection. Together, these structurally different encoders and model-specific gating networks implement model-based augmentation, producing complementary input views tailored to each architecture. The two encoders, trained with a shared decoder and cross-reconstruction loss, learn distinct yet aligned representations that reflect their respective inductive biases. During training, the OSDT encoder (robust to noise and effective at modeling localized, high-frequency structure) guides the neural encoder toward representations more aligned with tabular data. At inference, only the neural encoder is used, preserving flexibility and SSL compatibility. Spectral analysis highlights the distinct inductive biases of each encoder. Our method achieves consistent gains in low-label classification and regression across diverse tabular datasets, outperforming deep and tree-based supervised baselines.

Low-rank gradient-based optimization methods have significantly improved memory efficiency during the training of large language models (LLMs), enabling operations within constrained hardware without sacrificing performance. However, these methods primarily emphasize memory savings, often overlooking potential acceleration in convergence due to their reliance on standard isotropic steepest descent techniques, which can perform suboptimally in the highly anisotropic landscapes typical of deep networks, particularly LLMs. In this paper, we propose SUMO (Subspace-Aware Moment-Orthogonalization), an optimizer that employs exact singular value decomposition (SVD) for moment orthogonalization within a dynamically adapted low-dimensional subspace, enabling norm-inducing steepest descent optimization steps. By explicitly aligning optimization steps with the spectral characteristics of the loss landscape, SUMO effectively mitigates approximation errors associated with commonly used methods like Newton-Schulz orthogonalization approximation. We theoretically establish an upper bound on these approximation errors, proving their dependence on the condition numbers of moments, conditions we analytically demonstrate are encountered during LLM training. Furthermore, we both theoretically and empirically illustrate that exact orthogonalization via SVD substantially improves convergence rates while reducing overall complexity. Empirical evaluations confirm that SUMO accelerates convergence, enhances stability, improves performance, and reduces memory requirements by up to 20\% compared to state-of-the-art methods.

The effectiveness of stochastic gradient descent (SGD) in neural network optimization is significantly influenced by stochastic gradient noise (SGN). Following the central limit theorem, SGN was initially described as Gaussian, but recently Simsekli et al (2019) demonstrated that the $S\alpha S$ Lévy distribution provides a better fit for the SGN. This assertion was purportedly debunked and rebounded to the Gaussian noise model that had been previously proposed. This study provides robust, comprehensive empirical evidence that SGN is heavy-tailed and is better represented by the $S\alpha S$ distribution. Our experiments include several datasets and multiple models, both discriminative and generative. Furthermore, we argue that different network parameters preserve distinct SGN properties. We develop a novel framework based on a Lévy-driven stochastic differential equation (SDE), where one-dimensional Lévy processes describe each parameter. This leads to a more accurate characterization of the dynamics of SGD around local minima. We use our framework to study SGD properties near local minima; these include the mean escape time and preferable exit directions.

We study the problem of performing face verification with an efficient neural model$f$. The efficiency of$f$stems from simplifying the face verification problem from an embedding nearest neighbor search into a binary problem; each user has its own neural network$f$. To allow information sharing between different individuals in the training set, we do not train$f$directly but instead generate the model weights using a hypernetwork$h$. This leads to the generation of a compact personalized model for face identification that can be deployed on edge devices. Key to the method's success is a novel way of generating hard negatives and carefully scheduling the training objectives. Our model leads to a substantially small$f$requiring only 23k parameters and 5M floating point operations (FLOPS). We use six face verification datasets to demonstrate that our method is on par or better than state-of-the-art models, with a significantly reduced number of parameters and computational burden. Furthermore, we perform an extensive ablation study to demonstrate the importance of each element in our method.

Classical methods for acoustic scene mapping require the estimation of the time difference of arrival (TDOA) between microphones. Unfortunately, TDOA estimation is very sensitive to reverberation and additive noise. We introduce an unsupervised data-driven approach that exploits the natural structure of the data. Toward this goal, we adapt the recently proposed local conformal autoencoder (LOCA) – an offline deep learning scheme for extracting standardized data coordinates from measurements. Our experimental setup includes a microphone array that measures the transmitted sound source, whose position is unknown, at multiple locations across the acoustic enclosure. We demonstrate that our proposed scheme learns an isometric representation of the microphones’ spatial locations and can perform extrapolation over new and unvisited regions. The performance of our method is evaluated using a series of realistic simulations and compared with a classical approach and other dimensionality-reduction schemes. We further assess reverberation's influence on our framework’s results and show that it demonstrates considerable robustness.

Feature selection is a crucial tool in machine learning and is widely applied across various scientific disciplines. Traditional supervised methods generally identify a universal set of informative features for the entire population. However, feature relevance often varies with context, while the context itself may not directly affect the outcome variable. Here, we propose a novel architecture for contextual feature selection where the subset of selected features is conditioned on the value of *context variables*. Our new approach, Conditional Stochastic Gates (c-STG), models the importance of features using conditional Bernoulli variables whose parameters are predicted based on contextual variables. We introduce a hypernetwork that maps context variables to feature selection parameters to learn the context-dependent gates along with a prediction model. We further present a theoretical analysis of our model, indicating that it can improve performance and flexibility over population-level methods in complex feature selection settings. Finally, we conduct an extensive benchmark using simulated and real-world datasets across multiple domains demonstrating that c-STG can lead to improved feature selection capabilities while enhancing prediction accuracy and interpretability.

Clustering is a fundamental learning task widely used as a first step in data analysis. For example, biologists use cluster assignments to analyze genome sequences, medical records, or images. Since downstream analysis is typically performed at the cluster level, practitioners seek reliable and interpretable clustering models. We propose a new deep-learning framework for general domain tabular data that predicts interpretable cluster assignments at the instance and cluster levels. First, we present a self-supervised procedure to identify the subset of the most informative features from each data point. Then, we design a model that predicts cluster assignments and a gate matrix that provides cluster-level feature selection. Overall, our model provides cluster assignments with an indication of the driving feature for each sample and each cluster. We show that the proposed method can reliably predict cluster assignments in biological, text, image, and physics tabular datasets. Furthermore, using previously proposed metrics, we verify that our model leads to interpretable results at a sample and cluster level. Our code is available on https://github.com/jsvir/idc.

Accurate detection of outliers is crucial for the success of numerous data analysis tasks. In this context, we propose the Probabilistic Robust AutoEncoder (PRAE) that can simultaneously remove outliers during training (transductive) and learn a mapping that can be used to detect outliers in new data (inductive). We first present the Robust AutoEncoder (RAE) objective that excludes outliers while including a subset of samples (inliers) that can be effectively reconstructed using an AutoEncoder (AE). RAE minimizes the autoencoder’s reconstruction error while incorporating as many samples as possible. This could be formulated via regularization by subtracting an $\ell_0$ norm, counting the number of selected samples from the reconstruction term. As this leads to an intractable combinatorial problem, we propose two probabilistic relaxations of RAE, which are differentiable and alleviate the need for a combinatorial search. We prove that the solution to the PRAE problem is equivalent to the solution of RAE. We then use synthetic data to demonstrate that PRAE can accurately remove outliers in various contamination levels. Finally, we show that using PRAE for outlier detection leads to state-of-the-art results for inductive and transductive outlier detection.

We propose a modified density estimation problem that is highly effective for detecting anomalies in tabular data. Our approach assumes that the density function is relatively stable (with lower variance) around normal samples. We have verified this hypothesis empirically using a wide range of real-world data. Then, we present a variance-stabilized density estimation problem for maximizing the likelihood of the observed samples while minimizing the variance of the density around normal samples. To obtain a reliable anomaly detector, we introduce a spectral ensemble of autoregressive models for learning the variance-stabilized distribution. We have conducted an extensive benchmark with 52 datasets, demonstrating that our method leads to state-of-the-art results while alleviating the need for data-specific hyperparameter tuning. Finally, we have used an ablation study to demonstrate the importance of each of the proposed components, followed by a stability analysis evaluating the robustness of our model.

We propose a novel voice activity detection (VAD) model in a low-resource environment. Our key idea is to model VAD as a denoising task and construct a network that is designed to identify nuisance features for a speech classification task. We train the model to simultaneously identify irrelevant features while predicting the type of speech event. Our model contains only 7.8K parameters, outperforms the previously proposed methods on the AVA-Speech evaluation set, and provides comparative results on the HAVIC dataset. We present its architecture, experimental results, and ablation study on the model’s components. We publish the code and the model1.

Multi-modal high throughput biological data presents a great scientific opportunity and a significant computational challenge. In multi-modal measurements, every sample is observed simultaneously by two or more sets of sensors. In such settings, many observed variables in both modalities are often nuisance and do not carry information about the phenomenon of interest. Here, we propose a multi-modal unsupervised feature selection framework: identifying informative variables based on coupled high-dimensional measurements. Our method is designed to identify features associated with two types of latent low-dimensional structures: (i) shared structures that govern the observations in both modalities, and (ii) differential structures that appear in only one modality. To that end, we propose two Laplacian-based scoring operators. We incorporate the scores with differentiable gates that mask nuisance features and enhance the accuracy of the structure captured by the graph Laplacian. The performance of the new scheme is illustrated using synthetic and real datasets, including an extended biological application to single-cell multi-omics.

Tabular datasets with low-sample-size or many variables are prevalent in biomedicine. Practitioners in this domain prefer linear or tree-based models over neural networks since the latter are harder to interpret and tend to overfit when applied to tabular datasets. To address these neural networks’ shortcomings, we propose an intrinsically interpretable network for heterogeneous biomedical data. We design a locally sparse neural network where the local sparsity is learned to identify the subset of most relevant features for each sample. This sample-specific sparsity is predicted via a gating network, which is trained in tandem with the prediction network. By forcing the model to select a subset of the most informative features for each sample, we reduce model overfitting in low-sample-size data and obtain an interpretable model. We demonstrate that our method outperforms state-of-the-art models when applied to synthetic or real-world biomedical datasets using extensive experiments. Furthermore, the proposed framework dramatically outperforms existing schemes when evaluating its interpretability capabilities. Finally, we demonstrate the applicability of our model to two important biomedical tasks: survival analysis and marker gene identification.

Scientific observations may consist of a large number of variables (features). Selecting a subset of meaningful features is often crucial for identifying patterns hidden in the ambient space. In this paper, we present a method for unsupervised feature selection, and we demonstrate its advantage in clustering, a common unsupervised task. We propose a differentiable loss that combines a graph Laplacian-based score that favors low-frequency features with a gating mechanism for removing nuisance features. Our method improves upon the naive graph Laplacian score by replacing it with a gated variant computed on a subset of low-frequency features. We identify this subset by learning the parameters of continuously relaxed Bernoulli variables, which gate the entire feature space. We mathematically motivate the proposed approach and demonstrate that it is crucial to compute the graph Laplacian on the gated inputs rather than on the full feature set in the high noise regime. Using several real-world examples, we demonstrate the efficacy and advantage of the proposed approach over leading baselines.

Feature selection problems have been extensively studied in the setting of linear estimation (e.g. LASSO), but less emphasis has been placed on feature selection for non-linear functions. In this study, we propose a method for feature selection in neural network estimation problems. The new procedure is based on probabilistic relaxation of the $\ell_0$ norm of features, or the count of the number of selected features. Our $\ell_0$-based regularization relies on a continuous relaxation of the Bernoulli distribution; such relaxation allows our model to learn the parameters of the approximate Bernoulli distributions via gradient descent. The proposed framework simultaneously learns either a nonlinear regression or classification function while selecting a small subset of features. We provide an information-theoretic justification for incorporating Bernoulli distribution into feature selection. Furthermore, we evaluate our method using synthetic and real-life data to demonstrate that our approach outperforms other commonly used methods in both predictive performance and feature selection.

Detection and discrimination of seismic events have important implications. Precise detection of earthquakes may help prevent collateral damage and even save lives. On the other hand, the ability to identify explosions reliably not only helps prevent false alarms but also is crucial for monitoring nuclear experiments. In this work, we present a method for automatic discrimination of seismic events. A neural network is trained to fuse information from multiple seismic channels into a correlated space. Then, we augment the minority class to avoid class imbalance. Finally, a tree-based classifier is used to estimate the nature of the suspected event. We apply the proposed approach to 1609 events collected in Israel and Jordan. Our framework demonstrates improved precision and recall scores.

We propose a new type of generative model for high-dimensional data that learns a manifold geometry of the data, rather than density, and can generate points evenly along this manifold. This is in contrast to existing generative models that represent data density, and are strongly affected by noise and other artifacts of data collection. We demonstrate how this approach corrects sampling biases and artifacts, thus improves several downstream data analysis tasks, such as clustering and classification. Finally, we demonstrate that this approach is especially useful in biology where, despite the advent of single-cell technologies, rare subpopulations and gene-interaction relationships are affected by biased sampling. We show that SUGAR can generate hypothetical populations, and it is able to reveal intrinsic patterns and mutual-information relationships between genes on a single-cell RNA sequencing dataset of hematopoiesis.

The problem of learning from seismic recordings has been studied for years. There is a growing interest of developing automatic mechanisms for identifying the properties of a seismic event. One main motivation is the ability to have a reliable identification of man-made explosions. The availability of multiple high-dimensional observations has increased the use of machine learning techniques in a variety of fields. In this paper, we propose to use a kernel-fusion-based dimensionality reduction framework for generating meaningful seismic representations from raw data. The proposed method is tested on 2023 events that were recorded in Israel and Jordan. The method achieves promising results in the classification of event type as well as the estimation of the event location. The proposed fusion and dimensionality reduction tools may be applied to other types of geophysical data.