arXiv research

Ridgeless ReLU networks interpolate datasets and extrapolate based on curvature signs.

Paper proposes a method to reduce hallucinations in diffusion models using Laplacian score sharpening.

New method uses neural networks to interpolate stellar atmospheres with high precision.

Simulating fluid flow in geological formations requires mesh generation, lithology mapping to the cells, and computing geometric properties such as normal vectors and volume of cells. The purpose of this research work is to compute and process the geometrical information required for performing numerical simulations in…

Machine learning is used to approximate density functionals. For the model problem of the kinetic energy of non-interacting fermions in 1d, mean absolute errors below 1 kcal/mol on test densities similar to the training set are reached with fewer than 100 training densities. A predictor identifies if a test density is …

FDN improves probabilistic regressors' adaptability to distribution shifts.

iSTFTNet2 improves iSTFTNet's speed and lightness with 1D-2D CNN.

Improved robustness of 1D CNNs for heart arrhythmia classification.

Split learning preserves privacy in 1D CNN models for detecting heart abnormalities.

A new deep learning benchmark reduces resource needs.

SKI speeds up Toeplitz Neural Networks by avoiding explicit decay bias and using frequency response.

Gradient descent trains shallow neural networks to approximate functions in 1D.

Study uses 1D-CNNs to forecast mortality in ELSA survey.

Study on the complexity of 1D ReLU neural networks, proving growth in linear regions.

Study shows how 3+1D cosmologies can evolve to de Sitter space under certain conditions.

PAC-Bayes framework fails on simple 1D linear classification task.

Uniqueness of 1D bi-Schrödinger flow proven from flat torus to compact space.

Soil texture is important for many environmental processes. In this paper, we study the classification of soil texture based on hyperspectral data. We develop and implement three 1-dimensional (1D) convolutional neural networks (CNN): the LucasCNN, the LucasResNet which contains an identity block as residual network, a…

Estimates reliability of nuclear fuel using advanced modeling techniques.

REGS samples from unnormalized distributions using gradient flow and neural networks.

Study shows inflation in 3+1D cosmologies with bounded scalar potential and specific symmetry.

Estimates latent positions in 1D torus from noisy pairwise affinities.

This study analyzes a non-orientable spacetime model in 1+1D quantum gravity.

Capsule networks excel in understanding spatial relationships in 2D data for vision related tasks. Even though they are not designed to capture 1D temporal relationships, with TimeCaps we demonstrate that given the ability, capsule networks excel in understanding temporal relationships. To this end, we generate capsule…

Topological Quantum Field Theories (TQFTs) pertinent to some emergent low energy phenomena of condensed matter lattice models in 2+1 and 3+1D are explored. Many of our field theories are highly-interacting without free quadratic analogs. Some of our bosonic TQFTs can be regarded as the continuum field theory formulatio…

Study no-arbitrage conditions in 1D diffusion markets with interest rates.

New method trains generative models without discriminators, improving stability and accuracy.

A simple block configures optimal kernel sizes for time series classification.

This paper proposes a novel kernel-based optimization scheme to handle tasks in the analysis, e.g., signal spectral estimation and single-channel source separation of 1D non-stationary oscillatory data. The key insight of our optimization scheme for reconstructing the time-frequency information is that when a nonparame…

This paper shows how forward rate interpolations are equivalent to discount factor interpolations in yield curve construction.

The paper proposes an ensemble of convolution-based methods for fault detection in gearboxes.

In this paper, we introduce an inclined curves according to parallel transport frame. Also, we define a vector field called Darboux vector field of an inclined curve in and we give a new characterization such as: "α: I \subset R \rightarrow E^4 is an inclined curve \Leftrightarrow k_1 \int k_1ds + k_2 \int \k_2 +k_3ds …

Paper proposes a faster RAE with sequence-aware encoding.

The author studies regions foliated by 1D families of functions and their applications.

MCNO learns PDE solution operators using Monte Carlo sampling.

The paper proposes a method for generating uniform interpolations on data manifolds.

This paper reconstructs equivalence classes in 1D neural networks.

Deep neural networks continue to show improved performance with increasing depth, an encouraging trend that implies an explosion in the possible permutations of network architectures and hyperparameters for which there is little intuitive guidance. To address this increasing complexity, we propose Evolutionary DEep Net…

New FX option interpolations impact implied volatilities.

Kernel interpolation is inconsistent for norms with smoothness above a constant.

Bayesian interpolants explain neural network inferences concisely.

This paper proposes a framework based on deep convolutional neural networks (CNNs) for automatic heart sound classification using short-segments of individual heart beats. We design a 1D-CNN that directly learns features from raw heart-sound signals, and a 2D-CNN that takes inputs of two- dimensional time-frequency fea…

Near-interpolating models grow norms quickly, affecting generalization.

This paper shows the susceptibility of spectrogram-based audio classifiers to adversarial attacks and the transferability of such attacks to audio waveforms. Some commonly used adversarial attacks to images have been applied to Mel-frequency and short-time Fourier transform spectrograms, and such perturbed spectrograms…

The paper improves interpolation in generative models by using specific base distributions.

Self attention mechanisms have become a key building block in many state-of-the-art language understanding models. In this paper, we show that the self attention operator can be formulated in terms of 1x1 convolution operations. Following this observation, we propose several novel operators: First, we introduce a 2D ve…

Deep neural networks can interpolate any dataset in the overparametrized regime.

Interpolation hurts robust generalization even without noise.