Research
On-device research index

arXiv research

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.

169,051 papers · 148 categories

Trend · papers per month

240481721961 · Jun 202019922001200920172026
48 results for numerical training

Neural ODEs' performance varies with numerical method, requiring adaptive step size control.

problem Neural ODEs' performance depends on the numerical method used during training.
method Proposes an adaptive step size control algorithm to ensure a valid ODE without increasing computational cost.
result Valid Neural ODEs require careful numerical method selection and step size adaptation.

Improves E2E ASR performance on numeric sequences with additional training data and denormalization.

problem Challenges in recognizing numeric sequences out-of-vocabulary in ASR systems.
method Uses text-to-speech for additional numeric training data and a small-footprint neural network for denormalization.
result Reduction of WER by up to a factor of 8 in the longest numeric sequences.

NGRC shows numerical instabilities with short lags and high-degree polynomials.

problem Numerical instabilities in NGRC feature matrix.
method Combining numerical linear algebra and dynamical systems theory, we study feature matrix conditioning. We evaluate different numerical algorithms for solving the regularized least-squares problem.
result SVD-based training achieves accurate forecasts without regularization, preferable for short lags and high-degree polynomials.

Deep neural networks struggle with numerical instability during training.

problem Numerical instability in gradient descent training of deep neural networks.
method Analysis of floating-point arithmetic and gradient descent in ReLU neural networks.
result It is highly unlikely for ReLU networks to maintain a superlinear number of affine pieces during training.

SINGD improves KFAC for memory-efficiency and stability in low-precision training.

problem Memory inefficiency and numerical instability of KFAC in low-precision training.
method Formulated inverse-free KFAC update and imposed structures in Kronecker factors.
result SINGD is memory-efficient and numerically robust, often outperforming AdamW in half precision.

Efficiently trains GMMs for streaming data with SGD, addressing local optima and numerical instabilities.

problem Local optima and numerical instabilities in training GMMs for high-dimensional streaming data.
method Stochastic Gradient Descent (SGD) with adaptive annealing and exponential-free approximation.
result SGD approach trains GMMs without k-means initialization and outperforms sEM for high-dimensional data.

GNMR controls runtime stability in low-precision language model training.

problem Efficient low-precision training faces numerical risks at specific operators.
method GNMR compares gradient norms to historical means, applying bounded recovery actions.
result GNMR preserves high-fidelity quality with sparse, budgeted recovery.

Better neural arithmetic logic units improve cell counting model generalization.

problem Neural networks struggle with high cell counts outside training data range.
method Introduced Neural Arithmetic Logic Units (NALU) for arithmetic operations in existing architectures.
result Improved cell counting accuracy for higher numeric ranges with better generalization.

xVal tokenizes numbers continuously for better scientific model training.

problem Lack of continuous numerical tokenization for scientific datasets in LLMs.
method xVal: Continuous numerical tokenization strategy.
result xVal outperforms other numerical tokenization methods on scientific datasets.

Proposes a method to train neural networks that solve differential equations faster.

problem Training neural networks that solve differential equations becomes computationally expensive.
method Introduces a differentiable surrogate for numerical solver time cost using higher-order derivatives.
result Trains models that are faster to solve while maintaining nearly the same accuracy.

New method uses tensor trains for efficient PDE approximation.

problem High-dimensional PDEs and the curse of dimensionality.
method Tensor trains and backward stochastic differential equations for parabolic PDEs.
result Achieves a favorable trade-off between accuracy and computational efficiency.

Paper introduces NumLLM for better financial text understanding with numeric variables.

problem Poor performance of existing financial large language models in numeric financial text.
method Constructed financial corpus, fine-tuned with LoRA modules, merged into foundation model.
result NumLLM achieves best performance on financial question-answering benchmark, especially with numeric questions.

Dynamic Vocabulary Pruning stabilizes LLM training by removing low-probability tokens.

problem Training Large Language Models (LLMs) with Reinforcement Learning (RL) causes numerical divergence between inference and training.
method Dynamic Vocabulary Pruning (DVP) constrains the RL objective to a safe vocabulary that excludes low-probability tokens.
result DVP stabilizes training by reducing systematic bias introduced by the extreme tail of the token distribution.

Paper proposes efficient training for normalizing flows in Boltzmann generators.

problem Training normalizing flows for Boltzmann generators is computationally challenging and unstable.
method Regression Training of Normalizing Flows (RegFlow) using 2\ell_2-regression.
result RegFlow enables efficient and stable training of normalizing flows for Boltzmann generators.

New estimators reduce variance in training variational autoencoders with discrete latent variables.

problem Training variational autoencoders with discrete latent variables requires efficient gradient estimation.
method Introduce ReinMax-Rao and ReinMax-CV estimators using Rao-Blackwellisation and control variates.
result Demonstrate superior performance on training variational autoencoders with discrete latent spaces.

New ODE solvers improve training efficiency and accuracy.

problem Training Neural ODEs requires efficient and accurate gradient calculation.
method Presented algebraically reversible ODE solvers that are time and memory efficient, calculate exact gradients, and are numerically stable.
result Reversible solvers strictly improve upon previous architectures in efficiency and accuracy.

EXODUS improves training of SNNs by stabilizing gradients and reducing complexity.

problem Training SNNs using BPTT is time-consuming and numerically unstable.
method EXODUS modifies SLAYER to account for neuron reset and uses IFT for correct gradient calculation, eliminating manual scaling.
result EXODUS achieves comparable or better performance than SLAYER, especially in tasks with temporal features.

This paper compares two methods for training neural ODEs in time-series regression and CNFs.

problem Training neural ODEs for time-series regression and CNFs efficiently.
method Discretize-Optimize (Disc-Opt) vs. Optimize-Discretize (Opt-Disc) approaches.
result Disc-Opt methods can achieve similar performance as Opt-Disc at inference with drastically reduced training costs.

Neural networks struggle with long sequences, but a new method improves their performance.

problem Neural networks struggle to generalize to longer sequences and unseen data.
method Proposed a learned conditional masking mechanism and binary encoding for numbers.
result Models can now generalize far outside their training range with near-perfect accuracy.

Develops a numerical algorithm for stochastic impulse control using regression surrogates.

problem Optimal impulse control in stochastic processes.
method Generates statistical surrogates for continuation and intervention functions, recursively trained over simulated state trajectories.
result Demonstrates flexibility and extensibility of the numerical scheme through case studies.

Enhances FM models for numerical features using function basis encoding.

problem Challenges in incorporating numerical features into FM variants.
method Encoding numerical features into a vector of function values for learning segmentized functions.
result Improves model accuracy by learning segmentized functions of numerical features.

Neural Galerkin schemes use active learning to solve high-dimensional equations.

problem Inaccurate function approximations in high dimensions with limited training data.
method Neural Galerkin schemes based on deep learning with active learning for high-dimensional PDEs.
result Active data collection improves the numerical solution of high-dimensional equations.

A hybrid model combines physics and machine learning to predict unknown processes in numerical models.

problem Unknown or poorly represented processes in numerical models of the Earth System.
method Combining a physical model with a neural-net trained on observations.
result The hybrid model accurately predicts unknown processes with high correlation (close to 1).

A new method computes Greeks for multi-asset options using tensor trains and Fourier transforms.

problem Efficient computation of Greeks for multi-asset options with high accuracy and low sample complexity.
method Tensor train (TT) representations of Fourier-based pricing functions, combined with numerical differentiation or analytical approaches.
result Significant speed-ups of up to 105imes10^{5} imes over Monte Carlo simulations while maintaining comparable accuracy.

Langevin algorithms enhance training of deep neural networks for stochastic control problems.

problem Training acceleration for deep neural networks in stochastic control problems.
method Application of Langevin algorithms to minimize the loss of deep neural networks in stochastic control problems.
result Langevin algorithms improve training on various stochastic control problems.

A new method for estimating uncertainties in neural ODEs without numerical integration.

problem Accurate estimation of predictive uncertainties in neural ODEs.
method Distributional Gradient Matching (DGM) algorithm that jointly trains a smoother and a dynamics model.
result Significantly more accurate predictions compared to traditional methods.

Proposes a method to solve deep neural networks' local minimum problem.

problem Local minimum problem in deep neural networks training.
method Transforms cross-entropy loss into risk-averse error criterion, adjusts RSI, and uses convexity region.
result Trained deep learning machine is expected to be inside a global minimum's attraction basin.

Adapts deep learning models trained on simulated images for use with real images.

problem Difficulty in training deep neural networks on large amounts of experimental data.
method Adversarial domain adaptation method to mitigate domain shift between simulated and experimental image data.
result Adversarial domain adaptation successfully mitigates domain shift and improves numerical observer performance.

Improved ANN-based Monte Carlo simulation for Higgs decay events.

problem Accurate simulation of Higgs boson decay events.
method Monte Carlo simulation using an Artificial Neural Network (ANN) with improved training algorithm.
result The ANN simulation of Higgs decay is within 0.7% of the true value and achieves 26% unweighting efficiency.

Study shows how numerical discretization affects reconstructions and parameter distributions in nano metrology.

problem Impact of numerical discretization on parameter reconstructions and model parameter distributions.
method Bayesian target vector optimization, finite element model, Gaussian process, stochastic machine learning surrogate models, Markov chain Monte Carlo sampler.
result Numerical discretization parameters impact the accuracy and distribution of reconstructed model parameters.

Deep learning upscales geologic models efficiently.

problem Upscaling large-scale geologic models for efficient simulation.
method Theory-guided convolutional neural network (TgCNN) trained to approximate hydraulic conductivity relationships.
result Deep learning method achieves equivalent upscaling accuracy to numerical methods but with significantly improved efficiency.