arXiv research

Improves score estimation for noised targets using known clean scores.

The paper provides theoretical guarantees for behavior cloning using generative models.

New auto-encoder handles varying noise levels without retraining.

A method merges two pretrained diffusion experts to improve image quality and likelihood.

We consider the problem of reconstructing a low rank matrix from noisy observations of a subset of its entries. This task has applications in statistical learning, computer vision, and signal processing. In these contexts, "noise" generically refers to any contribution to the data that is not captured by the low-rank m…

Improved image quality in diffusion models by limiting guidance to a specific noise level range.

Image denoising techniques are essential to reducing noise levels and enhancing diagnosis reliability in low-dose computed tomography (CT). Machine learning based denoising methods have shown great potential in removing the complex and spatial-variant noises in CT images. However, some residue artifacts would appear in…

Classical convolutional neural networks (cCNNs) are very good at categorizing objects in images. But, unlike human vision which is relatively robust to noise in images, the performance of cCNNs declines quickly as image quality worsens. Here we propose to use recurrent connections within the convolutional layers to mak…

This paper studies least-square regression penalized with partly smooth convex regularizers. This class of functions is very large and versatile allowing to promote solutions conforming to some notion of low-complexity. Indeed, they force solutions of variational problems to belong to a low-dimensional manifold (the so…

We consider the problem of learning the level set for which a noisy black-box function exceeds a given threshold. To efficiently reconstruct the level set, we investigate Gaussian process (GP) metamodels. Our focus is on strongly stochastic samplers, in particular with heavy-tailed simulation noise and low signal-to-no…

This paper sets fundamental limits for rank-one matrix estimation with varying noise levels.

Vote-boosting is a sequential ensemble learning method in which the individual classifiers are built on different weighted versions of the training data. To build a new classifier, the weight of each training instance is determined in terms of the degree of disagreement among the current ensemble predictions for that i…

New method calibrates noise for attack risk, improving ML model accuracy.

FCNv2 robustness tested under noise and random initial conditions.

We introduce a new generative model where samples are produced via Langevin dynamics using gradients of the data distribution estimated with score matching. Because gradients can be ill-defined and hard to estimate when the data resides on low-dimensional manifolds, we perturb the data with different levels of Gaussian…

CutMix enhances feature learning in neural networks, improving test accuracy.

BDDMs eliminate noise conditioning in diffusion models, simplifying training and sampling.

Paper tackles robust matrix completion with heavy-tailed noise.

Study assesses CNN model robustness to noise in low-cost CT scans.

Enhanced consistency bounds derived for classification under a new noise condition.

Deep convolutional networks often append additive constant ("bias") terms to their convolution operations, enabling a richer repertoire of functional mappings. Biases are also used to facilitate training, by subtracting mean response over batches of training images (a component of "batch normalization"). Recent state-o…

The high-dimensional linear model $y = X β^0 + ε$ is considered and the focus is put on the problem of recovering the support $S^0$ of the sparse vector $β^0.$ We introduce Lasso-Zero, a new $\ell_1$-based estimator whose novelty resides in an "overfit, then threshold" paradigm and the use of noise dictionaries concate…

Image reconstruction in low-count PET is particularly challenging because gammas from natural radioactivity in Lu-based crystals cause high random fractions that lower the measurement signal-to-noise-ratio (SNR). In model-based image reconstruction (MBIR), using more iterations of an unregularized method may increase t…

LoRA and privacy: Random projections help but not always.

The paper introduces a frequency-domain estimator for low-order systems from noisy data.

The data processing inequality doesn't always hold in practice, showing benefits in low-level tasks.

We study the impact of input noise dimension on GAN performance.

Using a recently developed method of noise level estimation that makes use of properties of the coarse grained-entropy we have analyzed the noise level for the Dow Jones index and a few stocks from the New York Stock Exchange. We have found that the noise level ranges from 40 to 80 percent of the signal variance. The c…

New CSC model extracts EEG signals with low noise sensitivity.

Time changes of noise level at Warsaw Stock Market are analyzed using a recently developed method basing on properties of the coarse grained entropy. The condition of the minimal noise level is used to build an efficient portfolio. Our noise level approach seems to be a much better tool for risk estimations than standa…

We present a simple noise-robust margin-based active learning algorithm to find homogeneous (passing the origin) linear separators and analyze its error convergence when labels are corrupted by noise. We show that when the imposed noise satisfies the Tsybakov low noise condition (Mammen, Tsybakov, and others 1999; Tsyb…

GRIP2 improves deep learning feature selection robustness in correlated and noisy data.

In coronary CT angiography, a series of CT images are taken at different levels of radiation dose during the examination. Although this reduces the total radiation dose, the image quality during the low-dose phases is significantly degraded. To address this problem, here we propose a novel semi-supervised learning tech…

Scaled sparse linear regression jointly estimates the regression coefficients and noise level in a linear model. It chooses an equilibrium with a sparse regression method by iteratively estimating the noise level via the mean residual square and scaling the penalty in proportion to the estimated noise level. The iterat…

We present a theoretical analysis of Maximum a Posteriori (MAP) sequence estimation for binary symmetric hidden Markov processes. We reduce the MAP estimation to the energy minimization of an appropriately defined Ising spin model, and focus on the performance of MAP as characterized by its accuracy and the number of s…

Chart autoencoders learn latent features preserving manifold topology and geometry, with robust denoising capabilities.

Recently, deep models have had considerable success in several tasks, especially with low-level representations. However, effective learning from sparse noisy samples is a major challenge in most deep models, especially in domains with structured representations. Inspired by the proven success of human guided machine l…

Deep learning tackles low-photon nanoscale holographic phase retrieval.

In this paper, we study the pooled data problem of identifying the labels associated with a large collection of items, based on a sequence of pooled tests revealing the counts of each label within the pool. In the noiseless setting, we identify an exact asymptotic threshold on the required number of tests with optimal …

This work connects diffusion models to power iteration, revealing how low frequencies emerge earlier.

New method learns from noisy data without knowing noise level.

Characterizes uncertainty in low-rank matrix completion with noisy data.

We study a seemingly unexpected and relatively less understood overfitting aspect of a fundamental tool in sparse linear modeling - best subset selection, which minimizes the residual sum of squares subject to a constraint on the number of nonzero coefficients. While the best subset selection procedure is often perceiv…

Study bounds noise level in linear regression with dependent data.

PIE-PINN estimates elastic properties from noisy, low-res displacement data.

This paper evaluates a method to improve representations using incomplete external evidence across tasks.

Proposes isotonic recalibration for insurance pricing to ensure auto-calibration under low signal-to-noise ratio.

Study on sparse recovery with mixed-quality data, establishing sample-size conditions.