Image de couverture de livre

(Almost) Smooth Sailing: Towards Numerical Stability of Neural Networks Through Differentiable Regularization of the Condition Number

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Nenov, Rossen, Haider, Daniel, Balazs, Peter
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Optimization and Control
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Smoothing the Edges: Smooth Optimization for Sparse Regularization using Hadamard Overparametrization
par: Kolb, Chris, et autres
Publié: (2023)

Implicit Regularization in Perturbed Deep Matrix Factorization: Spectral Conditions and Stability
par: Wang, Jingzhe, et autres
Publié: (2026)

Smoothing Methods for Automatic Differentiation Across Conditional Branches
par: Kreikemeyer, Justin N., et autres
Publié: (2023)

Enhancing Stability of Physics-Informed Neural Network Training Through Saddle-Point Reformulation
par: Bylinkin, Dmitry, et autres
Publié: (2025)

Fixed-Point Automatic Differentiation of Forward--Backward Splitting Algorithms for Partly Smooth Functions
par: Mehmood, Sheheryar, et autres
Publié: (2022)

Physics-Informed Graph Neural Network for Dynamic Reconfiguration of Power Systems
par: Authier, Jules, et autres
Publié: (2023)

Efficiently Escaping Saddle Points under Generalized Smoothness via Self-Bounding Regularity
par: Cao, Daniel Yiming, et autres
Publié: (2025)

Almost Sure Convergence Analysis of Differentially Private Stochastic Gradient Methods
par: Mukherjee, Amartya, et autres
Publié: (2025)

Regularized Gauss-Newton for Optimizing Overparameterized Neural Networks
par: Adeoye, Adeyemi D., et autres
Publié: (2024)

Fair Supervised Learning Through Constraints on Smooth Nonconvex Unfairness-Measure Surrogates
par: Khatti, Zahra, et autres
Publié: (2025)

Stability Regularized Cross-Validation
par: Cory-Wright, Ryan, et autres
Publié: (2025)

Stability and Performance Analysis of Discrete-Time ReLU Recurrent Neural Networks
par: Noori, Sahel Vahedi, et autres
Publié: (2024)

How Neural Networks Learn the Support is an Implicit Regularization Effect of SGD
par: Beneventano, Pierfrancesco, et autres
Publié: (2024)

Regularized Gradient Clipping Provably Trains Wide and Deep Neural Networks
par: Tucat, Matteo, et autres
Publié: (2024)

Neural Network Training Techniques Regularize Optimization Trajectory: An Empirical Study
par: Chen, Cheng, et autres
Publié: (2020)

Why Smooth Stability Assumptions Fail for ReLU Learning
par: Katende, Ronald
Publié: (2025)

On the Complexity of Finite-Sum Smooth Optimization under the Polyak-Łojasiewicz Condition
par: Bai, Yunyan, et autres
Publié: (2024)

Towards Quantifying the Hessian Structure of Neural Networks
par: Dong, Zhaorui, et autres
Publié: (2025)

A Study of Condition Numbers for First-Order Optimization
par: Guille-Escuret, Charles, et autres
Publié: (2020)

Stability of Data-Dependent Ridge-Regularization for Inverse Problems
par: Neumayer, Sebastian, et autres
Publié: (2024)

Distributed Control of Network Systems in the Space of Stabilizing Graph Neural Network Policies
par: Cao, John, et autres
Publié: (2025)

Regularized Adaptive Momentum Dual Averaging with an Efficient Inexact Subproblem Solver for Training Structured Neural Network
par: Huang, Zih-Syuan, et autres
Publié: (2024)

An Adaptive and Stability-Promoting Layerwise Training Approach for Sparse Deep Neural Network Architecture
par: Krishnanunni, C G, et autres
Publié: (2022)

Differentiation Through Black-Box Quadratic Programming Solvers
par: Magoon, Connor W., et autres
Publié: (2024)

ATE-SG: Alternate Through the Epochs Stochastic Gradient for Multi-Task Neural Networks
par: Bellavia, Stefania, et autres
Publié: (2023)

Towards Guided Descent: Optimization Algorithms for Training Neural Networks At Scale
par: Nagwekar, Ansh
Publié: (2025)

A Unified Lyapunov-IQC Framework for Uniform Stability of Smooth Quadratic First-Order Accelerated Optimizers
par: Li, Don, et autres
Publié: (2026)

Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse
par: Jacot, Arthur, et autres
Publié: (2024)

Almost Surely $\sqrt{T}$ Regret for Adaptive LQR
par: Lu, Yiwen, et autres
Publié: (2023)

A Unified Theory of Stochastic Proximal Point Methods without Smoothness
par: Richtárik, Peter, et autres
Publié: (2024)

Towards Optimal Branching of Linear and Semidefinite Relaxations for Neural Network Robustness Certification
par: Anderson, Brendon G., et autres
Publié: (2021)

Towards Understanding Gradient Flow Dynamics of Homogeneous Neural Networks Beyond the Origin
par: Kumar, Akshay, et autres
Publié: (2025)

A full splitting algorithm for structured difference-of-convex programs
par: Bot, Radu Ioan, et autres
Publié: (2025)

Regularized Q-learning through Robust Averaging
par: Schmitt-Förster, Peter, et autres
Publié: (2024)

Recent Advances in Non-convex Smoothness Conditions and Applicability to Deep Linear Neural Networks
par: Patel, Vivak, et autres
Publié: (2024)

On the Condition Number Dependency in Bilevel Optimization
par: Chen, Lesi, et autres
Publié: (2025)

Kernel Learning with Adversarial Features: Numerical Efficiency and Adaptive Regularization
par: Ribeiro, Antônio H., et autres
Publié: (2025)

A Penalty Approach for Differentiation Through Black-Box Quadratic Programming Solvers
par: Linghu, Yuxuan, et autres
Publié: (2026)

Differentiable Convex Optimization Layers in Neural Architectures: Foundations and Perspectives
par: Katyal, Calder
Publié: (2024)

Control and optimization for Neural Partial Differential Equations in Supervised Learning
par: Bensoussan, Alain, et autres
Publié: (2025)