Image de couverture de livre

Ghosts of Softmax: Complex Singularities That Limit Safe Step Sizes in Cross-Entropy

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Sao, Piyush
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence 68T07, 90C26, 41A58 I.2.6; G.1.6
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

The Non-Linearity Perturbation Threshold: Width Scaling and Landscape Bifurcations in Deep Learning
par: Alexander, Michael
Publié: (2026)

CAO: Curvature-Adaptive Optimization via Periodic Low-Rank Hessian Sketching
par: Du, Wenzhang
Publié: (2025)

Differentiable Optimization Layers for Guaranteed Fairness in Deep Learning
par: Troxell, David, et autres
Publié: (2026)

GDNSQ: Gradual Differentiable Noise Scale Quantization for Low-bit Neural Networks
par: Salishev, Sergey, et autres
Publié: (2025)

Multi-Objective Optimization with Desirability and Morris-Mitchell Criterion
par: Bartz-Beielstein, Thomas, et autres
Publié: (2025)

Local properties of neural networks through the lens of layer-wise Hessians
par: Bolshim, Maxim, et autres
Publié: (2025)

Inter-Layer Hessian Analysis of Neural Networks with DAG Architectures
par: Bolshim, Maxim, et autres
Publié: (2026)

Stochastic Estimation of the Layer-wise Hessian Trace for Monitoring Neural-network Training
par: Bolshim, Maxim, et autres
Publié: (2026)

Complex-valued convolutional neural network classification of hand gesture from radar images
par: Khandan, Shokooh
Publié: (2024)

Limitations of Scalarisation in MORL: A Comparative Study in Discrete Environments
par: Shah, Muhammad Sa'ood, et autres
Publié: (2025)

Maximum Entropy Relaxation of Multi-Way Cardinality Constraints for Synthetic Population Generation
par: Pachet, François, et autres
Publié: (2026)

Refining Graphical Neural Network Predictions Using Flow Matching for Optimal Power Flow with Constraint-Satisfaction Guarantee
par: Khanal, Kshitiz
Publié: (2025)

Stable and Convexified Information Bottleneck Optimization via Symbolic Continuation and Entropy-Regularized Trajectories
par: Alpay, Faruk
Publié: (2025)

Bed-Attached Vibration Sensor System: A Machine Learning Approach for Fall Detection in Nursing Homes
par: Bartz-Beielstein, Thomas, et autres
Publié: (2024)

Simplifying Hyperparameter Tuning in Online Machine Learning -- The spotRiverGUI
par: Bartz-Beielstein, Thomas
Publié: (2024)

IGT-OMD: Implicit Gradient Transport for Decision-Focused Learning under Delayed Feedback
par: Amoh, Benjamin, et autres
Publié: (2026)

PyEPO: A PyTorch-based End-to-End Predict-then-Optimize Library for Linear and Integer Programming
par: Tang, Bo, et autres
Publié: (2022)

Theoretical Framework for Tempered Fractional Gradient Descent: Application to Breast Cancer Classification
par: Naifar, Omar
Publié: (2025)

NeurOptimisation: The Spiking Way to Evolve
par: Cruz-Duarte, Jorge Mario, et autres
Publié: (2025)

Interior-Point Vanishing Problem in Semidefinite Relaxations for Neural Network Verification
par: Ueda, Ryota, et autres
Publié: (2025)

On the Convergence Behavior of Preconditioned Gradient Descent Toward the Rich Learning Regime
par: Jiang, Shuai, et autres
Publié: (2026)

FlowAdam: Implicit Regularization via Geometry-Aware Soft Momentum Injection
par: Singh, Devender, et autres
Publié: (2026)

Multi-Objective Optimization and Hyperparameter Tuning With Desirability Functions
par: Bartz-Beielstein, Thomas
Publié: (2025)

Physics Informed Differentiable Solvers for Learning Parametric Solution Manifolds in Heterogeneous Physical Systems
par: Panahi, Milad, et autres
Publié: (2026)

Enhancing Model Based Derivative Free Optimization using Direct Search
par: Li, Zijun, et autres
Publié: (2026)

Benchmarking Generative AI Against Bayesian Optimization for Constrained Multi-Objective Inverse Design
par: Awan, Muhammad Bilal, et autres
Publié: (2025)

An in-depth look at approximation via deep and narrow neural networks
par: Dommel, Joris, et autres
Publié: (2025)

VORT: Adaptive Power-Law Memory for NLP Transformers
par: Mlaiki, Nabil
Publié: (2026)

Machine Learning Algorithms for Improving Black Box Optimization Solvers
par: Kimiaei, Morteza, et autres
Publié: (2025)

i-DEQ: A stable inertial deep equilibrium model for image restoration
par: Clerc, Antonin, et autres
Publié: (2026)

Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization
par: Vary, Simon, et autres
Publié: (2024)

Escaping Saddle Points via Curvature-Calibrated Perturbations: A Complete Analysis with Explicit Constants and Empirical Validation
par: Alpay, Faruk, et autres
Publié: (2025)

From Non-Identifiability to Goal-Integrated Decision-Making in Parametric Inverse Optimization
par: Ahmadi, Farzin, et autres
Publié: (2026)

TOPSIS-like metaheuristic for LABS problem
par: Urbańczyk, Aleksandra, et autres
Publié: (2025)

J6: Jacobian-Driven Role Attribution for Multi-Objective Prompt Optimization in LLMs
par: Wu, Yao
Publié: (2025)

Implicit Bias and Invariance: How Hopfield Networks Efficiently Learn Graph Orbits
par: Murray, Michael, et autres
Publié: (2025)

Setwise Coordinate Descent for Dual Asynchronous Decentralized Optimization
par: Costantini, Marina, et autres
Publié: (2025)

A unified convergence theory for adaptive first-order methods in the nonconvex case, including AdaNorm, full and diagonal AdaGrad, Shampoo and Muo
par: Gratton, S., et autres
Publié: (2026)

Adaptive Conditional Forest Sampling for Spectral Risk Optimisation under Decision-Dependent Uncertainty
par: Kurbucz, Marcell T.
Publié: (2026)

Pseudoconvex Problems in Operational Decision Systems: Algorithms for Joint Learning and Optimization
par: Li, Zijun, et autres
Publié: (2026)