Cover Image

Non-normal spectral signatures of instability in neural network training dynamics

Fuente: arXiv

Saved in:

Bibliographic Details
Main Author:	Ghosh, Souvik
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Disordered Systems and Neural Networks Materials Science Optimization and Control
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

High-Dimensional Analysis of Gradient Flow for Extensive-Width Quadratic Neural Networks
by: Martin, Simon, et al.
Published: (2026)

Asymptotic Dynamics of Alternating Minimization for Bilinear Regression
by: Okajima, Koki, et al.
Published: (2024)

Invariant multiscale neural networks for data-scarce scientific applications
by: Schurov, I., et al.
Published: (2024)

A Differentiable Measure of Algebraic Complexity: Provably Exact Discovery of Group Structures
by: Huh, Dongsung, et al.
Published: (2025)

CLuP practically achieves $\sim 1.77$ positive and $\sim 0.33$ negative Hopfield model ground state free energy
by: Stojnic, Mihailo
Published: (2025)

A CLuP algorithm to practically achieve $\sim 0.76$ SK--model ground state free energy
by: Stojnic, Mihailo
Published: (2025)

How does training shape the Riemannian geometry of neural network representations?
by: Zavatone-Veth, Jacob A., et al.
Published: (2023)

Competition, stability, and functionality in excitatory-inhibitory neural circuits
by: Betteti, Simone, et al.
Published: (2025)

Phonon predictions with E(3)-equivariant graph neural networks
by: Fang, Shiang, et al.
Published: (2024)

Representing Born effective charges with equivariant graph convolutional neural networks
by: Kutana, Alex, et al.
Published: (2024)

AiiDA-TrainsPot: Towards automated training of neural-network interatomic potentials
by: Bidoggia, Davide, et al.
Published: (2025)

Anisotropic Crystallization Kinetics and Interfacial Dynamics of Phase-Change Material Sb$_2$S$_3$ from Machine Learning Force Field Simulations
by: Chakraborty, Souvik, et al.
Published: (2026)

Interpretation of Crystal Energy Landscapes with Kolmogorov-Arnold Networks
by: Zu, Gen, et al.
Published: (2026)

Generative Inversion of Spectroscopic Data for Amorphous Structure Elucidation
by: Guo, Jiawei, et al.
Published: (2026)

STEM Diffraction Pattern Analysis with Deep Learning Networks
by: Wissel, Sebastian, et al.
Published: (2025)

Predicting and Interpreting Energy Barriers of Metallic Glasses with Graph Neural Networks
by: Li, Haoyu, et al.
Published: (2023)

Improving Electrolyte Performance for Target Cathode Loading Using Interpretable Data-Driven Approach
by: Sharma, Vidushi, et al.
Published: (2024)

Deep Generative Learning of Magnetic Frustration in Artificial Spin Ice from Magnetic Force Microscopy Images
by: Neogi, Arnab, et al.
Published: (2025)

Molecular Identification from AFM images using the IUPAC Nomenclature and Attribute Multimodal Recurrent Neural Networks
by: Carracedo-Cosme, Jaime, et al.
Published: (2022)

Bayesian Optimization of Multi-Bit Pulse Encoding in In2O3/Al2O3 Thin-film Transistors for Temporal Data Processing
by: Meza-Arroyo, Javier, et al.
Published: (2025)

Direct comparison of stochastic driven nonlinear dynamical systems for combinatorial optimization
by: Hou, Junpeng, et al.
Published: (2025)

Emergent weight morphologies in deep neural networks
by: de Jong, Pascal, et al.
Published: (2025)

High-dimensional learning of narrow neural networks
by: Cui, Hugo
Published: (2024)

Constructing material network representations for intelligent amorphous alloys design
by: Zhang, S. -Y., et al.
Published: (2025)

Optimization Algorithms for Multi-Species Spherical Spin Glasses
by: Huang, Brice, et al.
Published: (2023)

Optimizing Mean Field Spin Glasses with External Field
by: Sellke, Mark
Published: (2021)

Supervised and Unsupervised protocols for hetero-associative neural networks
by: Alessandrelli, Andrea, et al.
Published: (2025)

Deep neural networks from the perspective of ergodic theory
by: Zhang, Fan
Published: (2023)

Computing frustration and near-monotonicity in deep neural networks
by: Wendin, Joel, et al.
Published: (2025)

Finite-time Lyapunov exponents of deep neural networks
by: Storm, L., et al.
Published: (2023)

Iterative quantum optimisation with a warm-started quantum state
by: Yuan, Haomu, et al.
Published: (2025)

Training neural networks with structured noise improves classification and generalization
by: Benedetti, Marco, et al.
Published: (2023)

Duality between controllability and observability for target control and estimation in networks
by: Montanari, Arthur N., et al.
Published: (2024)

Boundary between noise and information applied to filtering neural network weight matrices
by: Staats, Max, et al.
Published: (2022)

Adaptive kernel predictors from feature-learning infinite limits of neural networks
by: Lauditi, Clarissa, et al.
Published: (2025)

Empirical universality and non-universality of local dynamics in the Sherrington-Kirkpatrick model
by: Liu, Grace, et al.
Published: (2025)

Non-equilibrium dynamics in geometrically frustrated spin glass Bi$_2$Fe$_3$GaO$_9$ with a Cairo lattice
by: Mihaylova, Desislava, et al.
Published: (2024)

Disorder-induced instability of a Weyl nodal loop semimetal towards a diffusive topological metal with protected multifractal surface states
by: Silva, João S., et al.
Published: (2024)

Statistical physics analysis of graph neural networks: Approaching optimality in the contextual stochastic block model
by: Duranthon, O., et al.
Published: (2025)

Properties of the geometry of solutions and capacity of multi-layer neural networks with Rectified Linear Units activations
by: Baldassi, Carlo, et al.
Published: (2019)