arXiv research

Spectral decoupling improves neural network generalization in medical imaging.

New method combines spectral and sparse methods for Gaussian processes.

Muon optimizer simplifies matrix optimization with spectral orthogonalization.

Muon optimizer improves deep learning with spectral norm constraints.

PolyNSD improves Neural Sheaf Diffusion with polynomial operators and spectral rescaling.

Inner product-based convolution has been a central component of convolutional neural networks (CNNs) and the key to learning visual representations. Inspired by the observation that CNN-learned features are naturally decoupled with the norm of features corresponding to the intra-class variation and the angle correspond…

The paper studies the convergence of harmonic metrics on Higgs bundles.

A spectral approach to building the exterior calculus in manifold learning problems is developed. The spectral approach is shown to converge to the true exterior calculus in the limit of large data. Simultaneously, the spectral approach decouples the memory requirements from the amount of data points and ambient space …

Defines and parametrizes $\mathfrak{sl}(2)$-type singular fibres in symplectic and odd orthogonal Hitchin systems.

The successive subspace learning (SSL) principle was developed and used to design an interpretable learning model, known as the PixelHop method,for image classification in our prior work. Here, we propose an improved PixelHop method and call it PixelHop++. First, to make the PixelHop model size smaller, we decouple a j…

Deep learning explained through spectral filtering of hierarchical features.

We investigate finite-time decoupled convergence in nonlinear two-time-scale stochastic approximation.

Decoupled GCN is shown to be equivalent to label propagation.

New gauge preserves Einstein metrics' interactions, proving rigidity on negatively curved manifolds.

A hybrid model combines diffusion and neural operator methods for stress prediction in hyperelastic materials.

OLS is a special case of Transformer, revealing its linear nature.

We investigate probabilistic decoupling of labels supplied for training, from the underlying classes for prediction. Decoupling enables an inference scheme general enough to implement many classification problems, including supervised, semi-supervised, positive-unlabelled, noisy-label and suggests a general solution to…

LASE learns graph embeddings by unrolling GD iterations into a neural network.

The paper proposes a decoupled approach to efficiently estimate CoVaR, a measure of systemic financial risk.

This paper investigates the effectiveness of decoupled weight decay at the start of training.

Clarifies method of phase synchronization for decoupling linear differential equations.

New insights explain why $β$-VAEs fail at disentanglement.

Kahler geometry explains decoupling of Kerr perturbations.

AdaDEM decouples EM into two parts to improve class overlap and uncertainty.

Recent variants improve knowledge distillation performance.

New algorithms for interpreting complex multivariate functions.

We establish decoupled functional CLTs for two-time-scale stochastic approximation.

Current reinforcement learning (RL) methods can successfully learn single tasks but often generalize poorly to modest perturbations in task domain or training procedure. In this work, we present a decoupled learning strategy for RL that creates a shared representation space where knowledge can be robustly transferred. …

Stylized facts can be regarded as constraints for any modeling attempt of price dynamics on a financial market, in that an empirically reasonable model has to reproduce these stylized facts at least qualitatively. The dynamics of market prices is modeled on a macro-level as the result of the dynamic coupling of two dyn…

Study spectral properties of sparse random graphs to recover latent vectors.

Backpropagation algorithm is indispensable for the training of feedforward neural networks. It requires propagating error gradients sequentially from the output layer all the way back to the input layer. The backward locking in backpropagation algorithm constrains us from updating network layers in parallel and fully l…

A new model decouples global and local image representations without supervision.

New equations simplify gauge-theoretic Khovanov homology solutions.

Decoupled PFNs improve sequential decision-making by separating epistemic and aleatoric uncertainties.

Novel Adam-family method with decoupled weight decay for training neural networks.

DAPS++ improves diffusion-based image restoration by decoupling prior and likelihood.

New method uses diffusion models for Bayesian inverse problems.

We investigate the convergence and stability properties of the decoupled extended Kalman filter learning algorithm (DEKF) within the long-short term memory network (LSTM) based online learning framework. For this purpose, we model DEKF as a perturbed extended Kalman filter and derive sufficient conditions for its stabi…

DAPS++ improves diffusion-based image restoration by decoupling prior and likelihood.

We study the convolutional phase retrieval problem, of recovering an unknown signal $\mathbf x \in \mathbb C^n$ from $m$ measurements consisting of the magnitude of its cyclic convolution with a given kernel $\mathbf a \in \mathbb C^m$. This model is motivated by applications such as channel estimation, optics, and u…

New analysis shows SNG's effectiveness in small samples.

FTPL policy achieves best-of-both-worlds regret in decoupled bandits with reduced computational cost.

New method detects heuristics in complex game strategies.

Introduces tunable basis functions for Gaussian processes.

Study compares constrained and decoupled moduli spaces of manifolds with particles and discs.

Unsupervised representation learning via generative modeling is a staple to many computer vision applications in the absence of labeled data. Variational Autoencoders (VAEs) are powerful generative models that learn representations useful for data generation. However, due to inherent challenges in the training objectiv…

The paper extends macroscopic market making to stochastic games, revealing properties and solving equations.

Unified q-learning for mean-field jump-diffusion models with unobservable population distribution.