A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
PaGoDA reduces diffusion model training costs by 64x.
Efficiently generates high-resolution images with reduced sampling time using LEGO bricks.
Researchers enhance EfficientNet models for practical efficiency on Graphcore IPU.
In this paper, we introduce a computational framework for recovering a high-resolution approximation of an unknown function from its low-resolution indirect measurements as well as high-resolution training observations by merging the frameworks of generalized sampling and functional principal component analysis. In par…
Deep learning improves 3D microscopy resolution without matched target images.
A cost-effective method to generate high-resolution images using wavelet-based super-resolution.
New deep learning method improves 4D Flow MRI super-resolution under domain shift.
A new multilevel framework speeds up ResNet training.
In this paper, we demonstrated a practical application of realistic river image generation using deep learning. Specifically, we explored a generative adversarial network (GAN) model capable of generating high-resolution and realistic river images that can be used to support modeling and analysis in surface water estim…
We propose a multi-level method to increase the accuracy of machine learning algorithms for approximating observables in scientific computing, particularly those that arise in systems modeled by differential equations. The algorithm relies on judiciously combining a large number of computationally cheap training data o…
Efficient and interpretable spatial analysis is crucial in many fields such as geology, sports, and climate science. Tensor latent factor models can describe higher-order correlations for spatial data. However, they are computationally expensive to train and are sensitive to initialization, leading to spatially incoher…
Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We …
CNN improves medium-range temperature forecasts with limited resources.
Training a deep neural network for classification constitutes a major problem in remote sensing due to the lack of adequate field data. Acquiring high-resolution ground truth (GT) by human interpretation is both cost-ineffective and inconsistent. We propose, instead, to utilize high-resolution, hyperspectral images for…
BayPOD-AL learns reduced-order models from high-fidelity data efficiently.
This research uses deep learning to automatically classify UN resolutions.
Image super-resolution (SR) is an underdetermined inverse problem, where a large number of plausible high-resolution images can explain the same downsampled image. Most current single image SR methods use empirical risk minimisation, often with a pixel-wise mean squared error (MSE) loss. However, the outputs from such …
A new deep learning method using Boolean logic reduces training and inference energy.
The `Folk Theorem' that a smooth action by a compact Lie group can be (canonically) resolved, by iterated blow up, to have unique isotropy type is proved in the context of manifolds with corners. This procedure is shown to capture the simultaneous resolution of all isotropy types in a `resolution tower' which projects …
Deep neural network models used for medical image segmentation are large because they are trained with high-resolution three-dimensional (3D) images. Graphics processing units (GPUs) are widely used to accelerate the trainings. However, the memory on a GPU is not large enough to train the models. A popular approach to …
WrapNet optimizes inference for low-resolution neural networks by using 8-bit additions.
Deep-learning improves 6x6-mm OCTA angiograms by reducing noise and artifacts.
Background: Three-dimensional, whole heart, balanced steady state free precession (WH-bSSFP) sequences provide delineation of intra-cardiac and vascular anatomy. However, they have long acquisition times. Here, we propose significant speed ups using a deep learning single volume super resolution reconstruction, to reco…
A scalable deep learning framework accelerates training of large neural networks for solving 3D Poisson equations.
A refined form of the `Folk Theorem' that a smooth action by a compact Lie group can be (canonically) resolved, by iterated blow up, to have unique isotropy type was established by the authors in the context of manifolds with corners; the canonical construction induces fibrations on the boundary faces of the resolution…
Neural networks have proven their capabilities by outperforming many other approaches on regression or classification tasks on various kinds of data. Other astonishing results have been achieved using neural nets as data generators, especially in settings of generative adversarial networks (GANs). One special applicati…
Many real-world phenomena are observed at multiple resolutions. Predictive models designed to predict these phenomena typically consider different resolutions separately. This approach might be limiting in applications where predictions are desired at fine resolutions but available training data is scarce. In this pape…
We reduce training time in convolutional networks (CNNs) with a method that, for some of the mini-batches: a) scales down the resolution of input images via downsampling, and b) reduces the forward pass operations via pooling on the convolution filters. Training is performed in an interleaved fashion; some batches unde…
We study the first-order scattering transform as a candidate for reducing the signal processed by a convolutional neural network (CNN). We show theoretical and empirical evidence that in the case of natural images and sufficiently small translation invariance, this transform preserves most of the signal information nee…
Recently, several models based on deep neural networks have achieved great success in terms of both reconstruction accuracy and computational performance for single image super-resolution. In these methods, the low resolution (LR) input image is upscaled to the high resolution (HR) space using a single filter, commonly…
ChemCPA predicts cellular responses to novel drugs using transfer learning.
Deep networks generalize well due to hidden mechanisms like renormalization.
FedSpace optimizes ML training on satellites and ground stations.
Generative adversarial networks generate realistic, time-evolving high-resolution atmospheric fields.
StreaMRAK improves KRR for streaming data.
MeshfreeFlowNet generates high-resolution spatio-temporal solutions from low-resolution inputs.
A second part of detailed elementary introduction into Khovanov homologies. This part is devoted to reduced Jones superpolynomials. The story is still about a hypercube of resolutions of a link diagram. Each resolution is a collection of non-intersecting cycles, and one associates a 2-dimensional vector space with each…
Geodesic rays of class C^{1,1} are constructed for any test configuration of a positive line bundle L on X using resolution of singularities. The construction reduces to finding a subsolution of the corresponding Monge-Ampere equation. Geometrically, this is accomplished by the use a positive line bundle on the resolut…
New method models PDEs from noisy, limited data.
RI-DeepONet learns neural operators from arbitrary sensor data.
EnScale learns to downscale climate models efficiently, capturing both spatial and temporal consistency.
We propose Progressive Structure-conditional Generative Adversarial Networks (PSGAN), a new framework that can generate full-body and high-resolution character images based on structural information. Recent progress in generative adversarial networks with progressive training has made it possible to generate high-resol…
Method fuses low and high-resolution data for better health estimates.
In our previous work we have shown that resistive cross point devices, so called Resistive Processing Unit (RPU) devices, can provide significant power and speed benefits when training deep fully connected networks as well as convolutional neural networks. In this work, we further extend the RPU concept for training re…
Deep generative models improve global precipitation forecasts.
For a data holder, such as a hospital or a government entity, who has a privately held collection of personal data, in which the revealing and/or processing of the personal identifiable data is restricted and prohibited by law. Then, "how can we ensure the data holder does conceal the identity of each individual in the…
Mapping the spatial distribution of poverty in developing countries remains an important and costly challenge. These "poverty maps" are key inputs for poverty targeting, public goods provision, political accountability, and impact evaluation, that are all the more important given the geographic dispersion of the remain…
Paper proposes combining GAM and DNN for accurate peak demand estimation from lower-resolution data.