arXiv subjects

Amir Averbuch

Publications and source records attributed to Amir Averbuch.

At least 19 recordsLinked to original sources

arXiv2026

The Singularity Space: A Generative Diffusion Framework for Signal Representation

Generative models often represent signals as dense grids of amplitudes, blurring sharp transients that are crucial for the correctness of physical signals. We introduce Singularity Space, a generative framework that represents signals through complex-plane singularities, rooted in the classical pole-residue representation of meromorphic functions. We learn a latent space of physically constrained, per-signal singularity configurations to solve an inverse problem from degraded or partial observations. The framework has three key properties: interpretability, in which each generated singularity configuration corresponds to a set of physical parameters; structural stability, which mitigates Gibbs artifacts at discontinuities; and resolution-free output reconstruction on arbitrary grids without retraining or interpolation. Our framework employs a transformer-based diffusion model that directly predicts samples at complex-plane singularity coordinates, subject to geometric constraints during sampling. As a controlled test case for sharp-feature recovery, we evaluate our framework on 1D Burgers shocks, where each shock is represented by 32 predicted singularities (an $8\times$ reduction versus a 1024-point grid signal). Our framework preserves signal structure ($\text{TV ratio} \approx 1$) under unseen test-time observation noise, achieves a $4.2\times$ lower reconstruction error in zero-shot sub-resolution generalization than a grid-based baseline, and recovers physical parameters to $10^{-4}$ absolute error in-distribution. These results suggest that singularity-based representations may provide a practical foundation for other transient-dominated signals such as speech and biomedical signals, with potential extension to higher-dimensional domains.

cs.LG↗

arXiv2024

SEL-CIE: Knowledge-Guided Self-Supervised Learning Framework for CIE-XYZ Reconstruction from Non-Linear sRGB Images

Modern cameras typically offer two types of image states: a minimally processed linear raw RGB image representing the raw sensor data, and a highly-processed non-linear image state, such as the sRGB state. The CIE-XYZ color space is a device-independent linear space used as part of the camera pipeline and can be helpful for computer vision tasks, such as image deblurring, dehazing, and color recognition tasks in medical applications, where color accuracy is important. However, images are usually saved in non-linear states, and achieving CIE-XYZ color images using conventional methods is not always possible. To tackle this issue, classical methodologies have been developed that focus on reversing the acquisition pipeline. More recently, supervised learning has been employed, using paired CIE-XYZ and sRGB representations of identical images. However, obtaining a large-scale dataset of CIE-XYZ and sRGB pairs can be challenging. To overcome this limitation and mitigate the reliance on large amounts of paired data, self-supervised learning (SSL) can be utilized as a substitute for relying solely on paired data. This paper proposes a framework for using SSL methods alongside paired data to reconstruct CIE-XYZ images and re-render sRGB images, outperforming existing approaches. The proposed framework is applied to the sRGB2XYZ dataset.

eess.IV↗

arXiv2023

TabADM: Unsupervised Tabular Anomaly Detection with Diffusion Models

Tables are an abundant form of data with use cases across all scientific fields. Real-world datasets often contain anomalous samples that can negatively affect downstream analysis. In this work, we only assume access to contaminated data and present a diffusion-based probabilistic model effective for unsupervised anomaly detection. Our model is trained to learn the density of normal samples by utilizing a unique rejection scheme to attenuate the influence of anomalies on the density estimation. At inference, we identify anomalies as samples in low-density regions. We use real data to demonstrate that our method improves detection capabilities over baselines. Furthermore, our method is relatively stable to the dimension of the data and does not require extensive hyperparameter tuning.

cs.LG↗

arXiv2023

Cross-boosting of WNNM Image Denoising method by Directional Wavelet Packets

The paper presents an image denoising scheme by combining a method that is based on directional quasi-analytic wavelet packets (qWPs) with the state-of-the-art Weighted Nuclear Norm Minimization (WNNM) denoising algorithm. The qWP-based denoising method (qWPdn) consists of multiscale qWP transform of the degraded image, application of adaptive localized soft thresholding to the transform coefficients using the Bivariate Shrinkage methodology, and restoration of the image from the thresholded coefficients from several decomposition levels. The combined method consists of several iterations of qWPdn and WNNM algorithms in a way that at each iteration the output from one algorithm boosts the input to the other. The proposed methodology couples the qWPdn capabilities to capture edges and fine texture patterns even in the severely corrupted images with utilizing the non-local self-similarity in real images that is inherent in the WNNM algorithm. Multiple experiments, which compared the proposed methodology with six advanced denoising algorithms, including WNNM, confirmed that the combined cross-boosting algorithm outperforms most of them in terms of both quantitative measure and visual perception quality.

eess.IV↗

arXiv2022

Imbalanced Classification via Explicit Gradient Learning From Augmented Data

Learning from imbalanced data is one of the most significant challenges in real-world classification tasks. In such cases, neural networks performance is substantially impaired due to preference towards the majority class. Existing approaches attempt to eliminate the bias through data re-sampling or re-weighting the loss in the learning process. Still, these methods tend to overfit the minority samples and perform poorly when the structure of the minority class is highly irregular. Here, we propose a novel deep meta-learning technique to augment a given imbalanced dataset with new minority instances. These additional data are incorporated in the classifier's deep-learning process, and their contributions are learned explicitly. The advantage of the proposed method is demonstrated on synthetic and real-world datasets with various imbalance ratios.

cs.LG↗

arXiv2022

Coupling BM3D with directional wavelet packets for image denoising

The paper presents an image denoising algorithm by combining a method that is based on directional quasi-analytic wavelet packets (qWPs) with the popular BM3D algorithm. The qWPs and its corresponding transforms are designed in [1]. The denoising algorithm qWP (qWPdn) applies an adaptive localized soft thresholding to the transform coefficients using the Bivariate Shrinkage methodology. The combined method consists of several iterations of qWPdn and BM3D algorithms, where the output from one algorithm updates the input to the other (cross-boosting).The qWPdn and BM3D methods complement each other. The qWPdn capabilities to capture edges and fine texture patterns are coupled with utilizing the sparsity in real images and self-similarity of patches in the image that is inherent in the BM3D. The obtained results are quite competitive with the best state-of-the-art algorithms. We compare the performance of the combined methodology with the performances of cptTP-CTF6, DAS-2 algorithms, which use directional frames, and the BM3D algorithm. In the overwhelming majority of the experiments, the combined algorithm outperformed the above methods.

eess.IV↗

arXiv2022

Imbalanced Classification via a Tabular Translation GAN

When presented with a binary classification problem where the data exhibits severe class imbalance, most standard predictive methods may fail to accurately model the minority class. We present a model based on Generative Adversarial Networks which uses additional regularization losses to map majority samples to corresponding synthetic minority samples. This translation mechanism encourages the synthesized samples to be close to the class boundary. Furthermore, we explore a selection criterion to retain the most useful of the synthesized samples. Experimental results using several downstream classifiers on a variety of tabular class-imbalanced datasets show that the proposed method improves average precision when compared to alternative re-weighting and oversampling techniques.

cs.LG↗

arXiv2021

Automated identification of transiting exoplanet candidates in NASA Transiting Exoplanets Survey Satellite (TESS) data with machine learning methods

A novel artificial intelligence (AI) technique that uses machine learning (ML) methodologies combines several algorithms, which were developed by ThetaRay, Inc., is applied to NASA's Transiting Exoplanets Survey Satellite (TESS) dataset to identify exoplanetary candidates. The AI/ML ThetaRay system is trained initially with Kepler exoplanetary data and validated with confirmed exoplanets before its application to TESS data. Existing and new features of the data, based on various observational parameters, are constructed and used in the AI/ML analysis by employing semi-supervised and unsupervised machine learning techniques. By the application of ThetaRay system to 10,803 light curves of threshold crossing events (TCEs) produced by the TESS mission, obtained from the Mikulski Archive for Space Telescopes, the algorithm yields about 50 targets for further analysis, and we uncover three new exoplanetary candidates by further manual vetting. This study demonstrates for the first time the successful application of the particular combined multiple AI/ML-based methodologies to a large astrophysical dataset for rapid automated classification of TCEs.

astro-ph.EP↗

arXiv2021

$\ell_0$-based Sparse Canonical Correlation Analysis

Canonical Correlation Analysis (CCA) models are powerful for studying the associations between two sets of variables. The canonically correlated representations, termed \textit{canonical variates} are widely used in unsupervised learning to analyze unlabeled multi-modal registered datasets. Despite their success, CCA models may break (or overfit) if the number of variables in either of the modalities exceeds the number of samples. Moreover, often a significant fraction of the variables measures modality-specific information, and thus removing them is beneficial for identifying the \textit{canonically correlated variates}. Here, we propose $\ell_0$-CCA, a method for learning correlated representations based on sparse subsets of variables from two observed modalities. Sparsity is obtained by multiplying the input variables by stochastic gates, whose parameters are learned together with the CCA weights via an $\ell_0$-regularized correlation loss. We further propose $\ell_0$-Deep CCA for solving the problem of non-linear sparse CCA by modeling the correlated representations using deep nets. We demonstrate the efficacy of the method using several synthetic and real examples. Most notably, by gating nuisance input variables, our approach improves the extracted representations compared to other linear, non-linear and sparse CCA-based models.

cs.LG↗

arXiv2021

Image inpainting using directional wavelet packets originating from polynomial splines

The paper presents a new algorithm for the image inpainting problem. The algorithm is using a recently designed versatile library of quasi-analytic complex-valued wavelet packets (qWPs) which originate from polynomial splines of arbitrary orders. Tensor products of 1D qWPs provide a diversity of 2D qWPs oriented in multiple directions. For example, a set of the fourth-level qWPs comprises 62 different directions. The properties of the presented qWPs such as refined frequency resolution, directionality of waveforms with unlimited number of orientations, (anti-)symmetry of waveforms and windowed oscillating structure of waveforms with a variety of frequencies, make them efficient in image processing applications, in particular, in dealing with the inpainting problem addressed in the paper. The obtained results for this problem are quite competitive with the best state-of-the-art algorithms. The inpainting is implemented by an iterative scheme, which, in essence, is the Split Bregman Iteration (SBI) procedure supplied with an adaptive variable soft thresholding based on the Bivariate Shrinkage algorithm. In the inpainting experiments, performance comparison between the qWP-based methods and the state-of-the-art algorithms is presented.

eess.IV↗

arXiv2020

Directional wavelet packets originating from polynomial splines

The paper presents a versatile library of quasi-analytic complex-valued wavelet packets (WPs) which originate from polynomial splines of arbitrary orders. The real parts of the quasi-analytic WPs are the regular spline-based orthonormal WPs designed in [1]. The imaginary parts are the so-called complementary orthonormal WPs that are derived from the Hilbert transforms of the regular WPs and, unlike the symmetric regular WPs, are antisymmetric. Tensor products of 1D quasi-analytic WPs provide a diversity of 2D WPs oriented in multiple directions. For example, a set of the fourth-level WPs comprises 62 different directions. The properties of the presented WPs are refined frequency resolution, directionality of waveforms with unlimited number of orientations, (anti-)symmetry of waveforms and windowed oscillating structure of waveforms with a variety of frequencies. Directional WPs have a strong potential to be used in various image processing applications such as restoration of degraded images and extraction of characteristic features from the images.

math.NA↗

arXiv2019

Analytic and directional wavelet packets in the space of periodic signals

The paper presents a versatile library of analytic and quasi-analytic complex-valued wavelet packets (WPs) which originate from discrete splines of arbitrary orders. The real parts of the quasi-analytic WPs are the regular spline-based orthonormal WPs designed in [2]. The imaginary parts are the so-called complementary orthonormal WPs, which, unlike the symmetric regular WPs, they are antisymmetric. Tensor products of 1D quasi-analytic WPs provide a diversity of 2D WPs oriented in multiple directions. For example, a set of the fourth-level WPs comprises 62 different directions. The designed computational scheme in the paper enables us to get fast and easy implementation of the WP transforms. The presented WPs proved to be efficient in signal/image processing applications such as restoration of images degraded by either additive noise or missing of up to 90% of their pixels.

math.NA↗

arXiv2019

MultiView Diffusion Maps

In this paper, we address the challenging task of achieving multi-view dimensionality reduction. The goal is to effectively use the availability of multiple views for extracting a coherent low-dimensional representation of the data. The proposed method exploits the intrinsic relation within each view, as well as the mutual relations between views. The multi-view dimensionality reduction is achieved by defining a cross-view model in which an implied random walk process is restrained to hop between objects in the different views. The method is robust to scaling and insensitive to small structural changes in the data. We define new diffusion distances and analyze the spectra of the proposed kernel. We show that the proposed framework is useful for various machine learning applications such as clustering, classification, and manifold learning. Finally, by fusing multi-sensor seismic data we present a method for automatic identification of seismic events.

cs.LG↗

arXiv2019

Kernel Scaling for Manifold Learning and Classification

Kernel methods play a critical role in many machine learning algorithms. They are useful in manifold learning, classification, clustering and other data analysis tasks. Setting the kernel's scale parameter, also referred to as the kernel's bandwidth, highly affects the performance of the task in hand. We propose to set a scale parameter that is tailored to one of two types of tasks: classification and manifold learning. For manifold learning, we seek a scale which is best at capturing the manifold's intrinsic dimension. For classification, we propose three methods for estimating the scale, which optimize the classification results in different senses. The proposed frameworks are simulated on artificial and on real datasets. The results show a high correlation between optimal classification rates and the estimated scales. Finally, we demonstrate the approach on a seismic event classification task.

cs.LG↗

arXiv2019

Multi-View Kernel Consensus For Data Analysis

The input data features set for many data driven tasks is high-dimensional while the intrinsic dimension of the data is low. Data analysis methods aim to uncover the underlying low dimensional structure imposed by the low dimensional hidden parameters by utilizing distance metrics that consider the set of attributes as a single monolithic set. However, the transformation of the low dimensional phenomena into the measured high dimensional observations might distort the distance metric, This distortion can effect the desired estimated low dimensional geometric structure. In this paper, we suggest to utilize the redundancy in the attribute domain by partitioning the attributes into multiple subsets we call views. The proposed methods utilize the agreement also called consensus between different views to extract valuable geometric information that unifies multiple views about the intrinsic relationships among several different observations. This unification enhances the information that a single view or a simple concatenations of views provides.

cs.LG↗

arXiv2018

Randomized LU decomposition: An Algorithm for Dictionaries Construction

In recent years, distinctive-dictionary construction has gained importance due to his usefulness in data processing. Usually, one or more dictionaries are constructed from a training data and then they are used to classify signals that did not participate in the training process. A new dictionary construction algorithm is introduced. It is based on a low-rank matrix factorization being achieved by the application of the randomized LU decomposition to a training data. This method is fast, scalable, parallelizable, consumes low memory, outperforms SVD in these categories and works also extremely well on large sparse matrices. In contrast to existing methods, the randomized LU decomposition constructs an under-complete dictionary, which simplifies both the construction and the classification processes of newly arrived signals. The dictionary construction is generic and general that fits different applications. We demonstrate the capabilities of this algorithm for file type identification, which is a fundamental task in digital security arena, performed nowadays for example by sandboxing mechanism, deep packet inspection, firewalls and anti-virus systems. We propose a content-based method that detects file types that neither depend on file extension nor on metadata. Such approach is harder to deceive and we show that only a few file fragments from a whole file are needed for a successful classification. Based on the constructed dictionaries, we show that the proposed method can effectively identify execution code fragments in PDF files. $\textbf{Keywords.}$ Dictionary construction, classification, LU decomposition, randomized LU decomposition, content-based file detection, computer security.

cs.CV↗

arXiv2017

Similarity Search Over Graphs Using Localized Spectral Analysis

This paper provides a new similarity detection algorithm. Given an input set of multi-dimensional data points, where each data point is assumed to be multi-dimensional, and an additional reference data point for similarity finding, the algorithm uses kernel method that embeds the data points into a low dimensional manifold. Unlike other kernel methods, which consider the entire data for the embedding, our method selects a specific set of kernel eigenvectors. The eigenvectors are chosen to separate between the data points and the reference data point so that similar data points can be easily identified as being distinct from most of the members in the dataset.

cs.AI↗

arXiv2017

Multi-View Kernels for Low-Dimensional Modeling of Seismic Events

The problem of learning from seismic recordings has been studied for years. There is a growing interest in developing automatic mechanisms for identifying the properties of a seismic event. One main motivation is the ability 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 work, 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 in Jordan. The method achieves promising results in classification of event type as well as in estimating the location of the event. The proposed fusion and dimensionality reduction tools may be applied to other types of geophysical data.

cs.LG↗