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.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

5 results for “autodifferentiation”

Reduced-Order Autodifferentiable Ensemble Kalman Filtersstat.ML

ROAD-EnKFs use learned low-dimensional models to improve state reconstruction and forecasting.

problem Reconstructing and forecasting states of unknown or expensive systems.

method Learned low-dimensional surrogate models and ensemble Kalman filter integration.

result ROAD-EnKFs achieve higher accuracy at lower computational cost than existing methods.

Kalman filters reduced-order modeling ensemble methods state reconstruction forecasting

2023-01-27abs ↗pdf ↗

A Quasilinear Algorithm for Computing Higher-Order Derivatives of Deep Feed-Forward Neural Networkscs.LG

A new algorithm speeds up neural network derivative calculations.

problem Exponential runtime of autodifferentiation for high-order derivatives in neural networks.

method n-TangentProp, a quasilinear algorithm for computing higher-order derivatives.

result Computes exact derivatives in quasilinear time, not exponential.

neural networks higher-order derivatives quasilinear time physics-informed neural networks autodifferentiation

2024-12-12abs ↗pdf ↗

Simple Linear Neuron Boostingcs.LG

Boosts linear neurons for faster convergence in neural networks.

problem Optimizing neural networks' neurons in function space.

method Reparameterizing network to whiten neuron features using Boosted Backpropagation.

result Equivalent to preconditioned gradient descent, leading to faster convergence.

neural networks boosted backpropagation preconditioning linear functions convergence

2025-02-03abs ↗pdf ↗

Exact Gradients for Stochastic Spiking Neural Networks Driven by Rough Signalsstat.ML

Framework for training stochastic spiking neural networks with rough signals.

problem Training stochastic spiking neural networks with noisy spike timing and dynamics.

method Rough path theory and signature kernels for gradient computation.

result Pathwise gradients of SSNNs' trajectories and event times exist and satisfy a recursive relation.

rough path theory stochastic spiking neural networks signature kernels gradient-based training càdlàg rough paths

2024-05-22abs ↗pdf ↗

Affine Normal Directions via Log-Determinant Geometry: Scalable Computation under Sparse Polynomial Structuremath.OC

New method computes affine normal directions efficiently for sparse polynomials.

problem Computing affine normal directions is computationally expensive in high dimensions.

method Reduces third-order tensor contraction to matrix-free formulation using log-determinant gradient.

result Scalable implementations with near-linear scaling in dimension and sparsity.

affine normal directions log-determinant geometry sparse polynomials scalable computation large-scale optimization

2026-04-01abs ↗pdf ↗