No article text was recorded for this version.
multi-layer convergence gap
Version 1 of 1185 papers192 conceptsfrozen 2026-10-04
Papers in this version
- Blog: A Geometric Approach of Gradient Descent Algorithms in Neural Networks
- Blog: Understanding Over-parameterization in Generative Adversarial Networks
- Blog: Global Convergence of Four-Layer Matrix Factorization under Random Initialization
- Blog: Gradient Descent on Infinitely Wide Neural Networks: Global Convergence and Generalization
- Blog: On the Convergence of Two-Layer Kolmogorov-Arnold Networks with First-Layer Training
- Blog: Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime
- Blog: Optimal and Diffusion Transports in Machine Learning
- Blog: Learning One-hidden-layer Neural Networks under General Input Distributions
- Blog: Overparameterized Nonlinear Learning: Gradient Descent Takes the Shortest Path?
- Blog: Learning Graph Neural Networks with Approximate Gradient Descent
- Blog: Can Shallow Neural Networks Beat the Curse of Dimensionality? A mean field training perspective
- Blog: Law of Large Numbers for Bayesian two-layer Neural Network trained with Variational Inference
- Blog: Full-Graph vs. Mini-Batch Training: Comprehensive Analysis from a Batch Size and Fan-Out Size Perspective
- Blog: How Much Pre-training Is Enough to Discover a Good Subnetwork?
- Blog: Application of Pretrained Deep Neural Networks to Large Vocabulary Speech Recognition
- Blog: On the Convergence of Shallow Neural Network Training with Randomly Masked Neurons
- Blog: The Implicit Bias of Batch Normalization in Linear Models and Two-layer Linear Convolutional Neural Networks
- Blog: A Mean-Field Analysis of Neural Stochastic Gradient Descent-Ascent for Functional Minimax Optimization
- Blog: Stochastic gradient descent for streaming linear and rectified linear systems with Massart noise
- Blog: DebiNet: Debiasing Linear Models with Nonlinear Overparameterized Neural Networks
- Blog: Neural Networks as Interacting Particle Systems: Asymptotic Convexity of the Loss Landscape and Universal Scaling of the Approximation Error
- Blog: A Dual Certificates Analysis of Compressive Off-the-Grid Recovery
- Blog: The Implicit Bias of Batch Normalization in Linear Models and Two-layer Linear Convolutional Neural Networks
- Blog: A proof of convergence for stochastic gradient descent in the training of artificial neural networks with ReLU activation for constant target functions
- Blog: Does a sparse ReLU network training problem always admit an optimum?
- Blog: Toward Understanding the Feature Learning Process of Self-supervised Contrastive Learning
- Blog: Gradient Descent Provably Optimizes Over-parameterized Neural Networks
- Blog: Understanding the Loss Surface of Single-Layered Neural Networks for Binary Classification
- Blog: Provable Non-linear Inductive Matrix Completion
- Blog: Recovery Guarantees for One-hidden-layer Neural Networks
- Blog: Physics-Informed Neural Networks: Minimizing Residual Loss with Wide Networks and Effective Activations
- Blog: The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?
- Blog: A Non-Parametric Regression Viewpoint : Generalization of Overparametrized Deep RELU Network Under Noisy Observations
- Blog: The inductive bias of ReLU networks on orthogonally separable data
- Blog: Global Convergence of SGD For Logistic Loss on Two Layer Neural Nets
- Blog: Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse
- Blog: Towards Understanding the Dynamics of the First-Order Adversaries
- Blog: Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks
- Blog: The Effects of Mild Over-parameterization on the Optimization Landscape of Shallow ReLU Neural Networks
- Blog: Topology and Geometry of Half-Rectified Network Optimization
- Blog: Global convergence of neuron birth-death dynamics
- Blog: AdaLoss: A Computationally-Efficient and Provably Convergent Adaptive Gradient Method
- Blog: Benefit of deep learning with non-convex noisy gradient descent: Provable excess risk bound and superiority to kernel methods
- Blog: Recovery Guarantees for One-hidden-layer Neural Networks
- Blog: Landscape Complexity for the Empirical Risk of Generalized Linear Models
- Blog: Leveraging the two-timescale regime to demonstrate convergence of neural networks
- Blog: GIST: Distributed Training for Large-Scale Graph Convolutional Networks
- Blog: Provably Efficient Neural GTD for Off-Policy Learning
- Blog: Tight Sample Complexity of Learning One-hidden-layer Convolutional Neural Networks
- Blog: The Noise Geometry of Stochastic Gradient Descent: A Quantitative and Analytical Characterization
- Blog: On the convergence of group-sparse autoencoders
- Blog: Solving Neural Min-Max Games: The Role of Architecture, Initialization & Dynamics
- Blog: On the Convergence of Certified Robust Training with Interval Bound Propagation
- Blog: On the validity of kernel approximations for orthogonally-initialized neural networks
- Blog: When is a Convolutional Filter Easy To Learn?
- Blog: Porcupine Neural Networks: (Almost) All Local Optima are Global
- Blog: On the Global Convergence of Gradient Descent for Over-parameterized Models using Optimal Transport
- Blog: Convergence of mean-field Langevin dynamics: Time and space discretization, stochastic gradient, and variance reduction
- Blog: Gradient Descent Provably Optimizes Over-parameterized Neural Networks
- Blog: Learning ReLUs via Gradient Descent
Showing 60 of 185 papers.
History & citation
Current · Permanent link
- This direction split from Global Convergence of Overparameterized Multi Layer Networks (85% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_multi_layer_convergence_gap_13920e45,
title = {multi-layer convergence gap},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/multi-layer-convergence-gap/rdr_13920e45868a4a959689f9ecbb16329e},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}This exact version (version 1, frozen text):
@misc{lacuna_multi_layer_convergence_gap_13920e45_v1,
title = {multi-layer convergence gap},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/multi-layer-convergence-gap/rdr_13920e45868a4a959689f9ecbb16329e/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-21},
}