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.

169,181 papers · 148 categories

Trend · papers per month

count % of corpus

7 results for “eigenspectra”

Hessian Eigenspectra of More Realistic Nonlinear Modelsstat.ML

Characterizes Hessian eigenspectra for realistic nonlinear models.

problem Understanding Hessian eigenspectra in realistic nonlinear models.

method Deterministic equivalent techniques from random matrix theory.

result Hessian can have qualitatively different spectral behaviors.

Hessian eigenspectra nonlinear models random matrix theory machine learning spectral behaviors

2021-03-02abs ↗pdf ↗

Does the Data Induce Capacity Control in Deep Learning?cs.LG

Data structure affects deep learning performance, study finds.

problem Understanding why deep learning performs poorly on typical datasets.

method Analyzed input correlation matrices and network Hessians, developed PAC-Bayes bounds.

result Sloppy eigenspectra in input data correlate with poor deep learning performance.

deep learning generalization PAC-Bayes sloppy eigenspectra input correlation

2021-10-27abs ↗pdf ↗

Characterization of catastrophic instabilities: Market crashes as paradigmq-fin.GN

Study reveals signatures of market crashes through eigenvalue analysis of stock return matrices.

problem Understanding the complexity and dynamics of market crashes.

method Cross-correlation structures and eigenspectra of stock return matrices were analyzed over different epochs.

result The smallest eigenvalue can distinguish between internal and external market instabilities.

market crashes eigenvalues cross-correlation financial instability eigenspectra

2018-01-22abs ↗pdf ↗

Evolutionary correlation, regime switching, spectral dynamics and optimal trading strategies for cryptocurrencies and equitiesq-fin.ST

The paper analyzes cryptocurrency and equity markets using advanced statistical methods.

problem Comparing dynamics and strategies between cryptocurrency and equity markets.

method Random matrix theory, PCA, spectral dynamics, structural break analysis, portfolio simulation.

result Cryptocurrency and equity markets exhibit distinct evolutionary dynamics and time-varying sector behaviors.

cryptocurrency equity market dynamics portfolio management spectral analysis

2021-12-31abs ↗pdf ↗

Non-normal Recurrent Neural Network (nnRNN): learning long time dependencies while improving expressivity with transient dynamicscs.LG

Proposes a new RNN structure to improve expressivity without sacrificing stability.

problem Exploding and vanishing gradient problems in RNNs and reduced expressivity.

method Introduces a non-normal RNN structure using Schur decomposition and splitting.

result Enhances expressivity while maintaining stability and training speed.

Recurrent Neural Networks Schur Decomposition Expressivity Stability Gradient Problems

2019-05-28abs ↗pdf ↗

Deep Equilibrium Models are Almost Equivalent to Not-so-deep Explicit Models for High-dimensional Gaussian Mixturescs.LG

DEQs and explicit networks are nearly equivalent for Gaussian mixtures.

problem Understanding the equivalence between DEQs and explicit neural networks.

method Random matrix theory and analysis of kernel matrices.

result A shallow explicit network can mimic the kernel of a DEQ.

Deep Equilibrium Models Implicit Neural Networks Explicit Neural Networks Gaussian Mixtures Random Matrix Theory

2024-02-05abs ↗pdf ↗

Spectral Estimation with Free Decompressionstat.ML

New method estimates large matrices' spectra from small sub-matrices.

problem Estimating large matrices' spectra when full matrix-vector products are not available.

method Free decompression based on free probability theory.

result Estimates eigenspectrum of impalpable matrices from small sub-matrices.

spectral estimation matrix spectrum free probability impalpable matrices random matrix theory

2025-06-13abs ↗pdf ↗