Imagen de Portada

Neural Collapse is Globally Optimal in Deep Regularized ResNets and Transformers

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Súkeník, Peter, Lampert, Christoph H., Mondelli, Marco
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Neural Collapse versus Low-rank Bias: Is Deep Neural Collapse Really Optimal?
por: Súkeník, Peter, et al.
Publicado: (2024)

Sink vs. diagonal patterns as mechanisms for attention switch and oversmoothing prevention
por: Súkeník, Peter, et al.
Publicado: (2026)

ResNets Are Deeper Than You Think
por: Mehmeti-Göpel, Christian H. X. Ali, et al.
Publicado: (2025)

Progressive Feedforward Collapse of ResNet Training
por: Wang, Sicong, et al.
Publicado: (2024)

L-Lipschitz Gershgorin ResNet Network
por: Juston, Marius F. R., et al.
Publicado: (2025)

Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse
por: Jacot, Arthur, et al.
Publicado: (2024)

Bridging Neural ODE and ResNet: A Formal Error Bound for Safety Verification
por: Sayed, Abdelrahman Sayed, et al.
Publicado: (2025)

On the inductive bias of infinite-depth ResNets and the bottleneck rank
por: Boix-Adsera, Enric
Publicado: (2025)

Hamiltonian Mechanics of Feature Learning: Bottleneck Structure in Leaky ResNets
por: Jacot, Arthur, et al.
Publicado: (2024)

Acute Lymphoblastic Leukemia Diagnosis Employing YOLOv11, YOLOv8, ResNet50, and Inception-ResNet-v2 Deep Learning Models
por: Awad, Alaa, et al.
Publicado: (2025)

Cardioformer: Advancing AI in ECG Analysis with Multi-Granularity Patching and ResNet
por: Mobin, Md Kamrujjaman, et al.
Publicado: (2025)

Intriguing Properties of Input-dependent Randomized Smoothing
por: Súkeník, Peter, et al.
Publicado: (2021)

A Neural Architecture Search Method using Auxiliary Evaluation Metric based on ResNet Architecture
por: Wang, Shang, et al.
Publicado: (2025)

Average gradient outer product as a mechanism for deep neural collapse
por: Beaglehole, Daniel, et al.
Publicado: (2024)

Transformative or Conservative? Conservation laws for ResNets and Transformers
por: Marcotte, Sibylle, et al.
Publicado: (2025)

Learning Quantized Continuous Controllers for Integer Hardware
por: Kresse, Fabian, et al.
Publicado: (2025)

PlantDiseaseNet-RT50: A Fine-tuned ResNet50 Architecture for High-Accuracy Plant Disease Detection Beyond Standard CNNs
por: Sagnika, Santwana, et al.
Publicado: (2025)

Lung Cancer Classification from CT Images Using ResNet
por: Adekunle, Olajumoke O., et al.
Publicado: (2025)

Generalization of Scaled Deep ResNets in the Mean-Field Regime
por: Chen, Yihang, et al.
Publicado: (2024)

ResNet: Enabling Deep Convolutional Neural Networks through Residual Learning
por: Liu, Xingyu, et al.
Publicado: (2025)

Towards an Optimal Control Perspective of ResNet Training
por: Püttschneider, Jens, et al.
Publicado: (2025)

Deep Neural Regression Collapse
por: Rangamani, Akshay, et al.
Publicado: (2026)

Optimal Regularization for Performative Learning
por: Cyffers, Edwige, et al.
Publicado: (2025)

OT-Transformer: A Continuous-time Transformer Architecture with Optimal Transport Regularization
por: Kan, Kelvin, et al.
Publicado: (2025)

Understanding Representation of Deep Equilibrium Models from Neural Collapse Perspective
por: Sun, Haixiang, et al.
Publicado: (2024)

Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime
por: Wu, Diyuan, et al.
Publicado: (2025)

MFI-ResNet: Efficient ResNet Architecture Optimization via MeanFlow Compression and Selective Incubation
por: Sun, Nuolin, et al.
Publicado: (2025)

ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videos
por: Dass, Sharana Dharshikgan Suresh, et al.
Publicado: (2024)

Scaling ResNets in the Large-depth Regime
por: Marion, Pierre, et al.
Publicado: (2022)

From Alexnet to Transformers: Measuring the Non-linearity of Deep Neural Networks with Affine Optimal Transport
por: Bouniot, Quentin, et al.
Publicado: (2023)

On the Design Space Between Transformers and Recursive Neural Nets
por: Chowdhury, Jishnu Ray, et al.
Publicado: (2024)

Lightweight Quantum-Enhanced ResNet for Coronary Angiography Classification: A Hybrid Quantum-Classical Feature Enhancement Framework
por: Xia, Jingsong
Publicado: (2026)

ENOT: Expectile Regularization for Fast and Accurate Training of Neural Optimal Transport
por: Buzun, Nazar, et al.
Publicado: (2024)

Transformers as Neural Operators for Solutions of Differential Equations with Finite Regularity
por: Shih, Benjamin, et al.
Publicado: (2024)

The Prevalence of Neural Collapse in Neural Multivariate Regression
por: Andriopoulos, George, et al.
Publicado: (2024)

Deep Learning Evidence for Global Optimality of Gerver's Sofa
por: Leng, Kuangdai, et al.
Publicado: (2024)

Avoiding Premature Collapse: Adaptive Annealing for Entropy-Regularized Structural Inference
por: Liu, Yizhi
Publicado: (2026)

Preventing Dimensional Collapse in Self-Supervised Learning via Orthogonality Regularization
por: He, Junlin, et al.
Publicado: (2024)

Collapsed Inference for Bayesian Deep Learning
por: Zeng, Zhe, et al.
Publicado: (2023)

AdaResNet: Enhancing Residual Networks with Dynamic Weight Adjustment for Improved Feature Integration
por: Su, Hong
Publicado: (2024)