English
Related papers

Related papers: Stabilize Deep ResNet with A Sharp Scaling Factor …

200 papers

A main puzzle of deep networks revolves around the absence of overfitting despite large overparametrization and despite the large capacity demonstrated by zero training error on randomly labeled data. In this note, we show that the dynamics…

A surprising phenomenon in the training of neural networks is the ability of gradient descent to find global minimizers of the training loss despite its non-convexity. Following earlier works, we investigate this behavior for wide shallow…

Optimization and Control · Mathematics 2026-05-12 Romain Petit , Clarice Poon , Gabriel Peyré

Recently, DeepNorm scales Transformers into extremely deep (i.e., 1000 layers) and reveals the promising potential of deep scaling. To stabilize the training of deep models, DeepNorm (Wang et al., 2022) attempts to constrain the model…

Machine Learning · Computer Science 2023-05-05 Yijin Liu , Xianfeng Zeng , Fandong Meng , Jie Zhou

In this paper, we theoretically prove that gradient descent can find a global minimum of non-convex optimization of all layers for nonlinear deep neural networks of sizes commonly encountered in practice. The theory developed in this paper…

Machine Learning · Statistics 2020-06-18 Kenji Kawaguchi , Jiaoyang Huang

Residual neural networks (ResNets) are a promising class of deep neural networks that have shown excellent performance for a number of learning tasks, e.g., image classification and recognition. Mathematically, ResNet architectures can be…

Optimization and Control · Mathematics 2019-07-26 S. Günther , L. Ruthotto , J. B. Schroder , E. C. Cyr , N. R. Gauger

We show that the gradient descent algorithm provides an implicit regularization effect in the learning of over-parameterized matrix factorization models and one-hidden-layer neural networks with quadratic activations. Concretely, we show…

Machine Learning · Computer Science 2019-02-15 Yuanzhi Li , Tengyu Ma , Hongyang Zhang

Recent advances have significantly improved our understanding of the generalization performance of gradient descent (GD) methods in deep neural networks. A natural and fundamental question is whether GD can achieve generalization rates…

Machine Learning · Computer Science 2026-04-14 Yuanfan Li , Yunwen Lei , Zheng-Chu Guo , Yiming Ying

The implicit bias induced by the training of neural networks has become a topic of rigorous study. In the limit of gradient flow and gradient descent with appropriate step size, it has been shown that when one trains a deep linear network…

Machine Learning · Computer Science 2022-04-27 Thien Le , Stefanie Jegelka

Overparametrization is a key factor in the absence of convexity to explain global convergence of gradient descent (GD) for neural networks. Beside the well studied lazy regime, infinite width (mean field) analysis has been developed for…

Neural and Evolutionary Computing · Computer Science 2023-02-07 Raphaël Barboni , Gabriel Peyré , François-Xavier Vialard

The stability and generalization of stochastic gradient-based methods provide valuable insights into understanding the algorithmic performance of machine learning models. As the main workhorse for deep learning, stochastic gradient descent…

Machine Learning · Statistics 2021-02-24 Tao Sun , Dongsheng Li , Bao Wang

We consider the problem of learning an unknown ReLU network with respect to Gaussian inputs and obtain the first nontrivial results for networks of depth more than two. We give an algorithm whose running time is a fixed polynomial in the…

Machine Learning · Computer Science 2020-09-29 Sitan Chen , Adam R. Klivans , Raghu Meka

Fine-tuning pretrained models has become a standard approach to adapting pretrained knowledge to improve the accuracy on new sparse, imbalance datasets. However, issues arise when optimization falls into a collapsed state, where the model…

Machine Learning · Computer Science 2026-05-01 Nghia Bui , Lijing Wang

Traditional analyses of gradient descent show that when the largest eigenvalue of the Hessian, also known as the sharpness $S(\theta)$, is bounded by $2/\eta$, training is "stable" and the training loss decreases monotonically. Recent…

Machine Learning · Computer Science 2023-04-12 Alex Damian , Eshaan Nichani , Jason D. Lee

Accelerated algorithms for minimizing smooth strongly convex functions usually require knowledge of the strong convexity parameter $\mu$. In the case of an unknown $\mu$, current adaptive techniques are based on restart schemes. When the…

Optimization and Control · Mathematics 2019-06-10 Mathieu Barré , Alexandre d'Aspremont

This work aims to provide understandings on the remarkable success of deep convolutional neural networks (CNNs) by theoretically analyzing their generalization performance and establishing optimization guarantees for gradient descent based…

Machine Learning · Computer Science 2018-05-29 Pan Zhou , Jiashi Feng

Effectively scaling up deep reinforcement learning models has proven notoriously difficult due to network pathologies during training, motivating various targeted interventions such as periodic reset and architectural advances such as layer…

Machine Learning · Computer Science 2025-06-23 Guozheng Ma , Lu Li , Zilin Wang , Li Shen , Pierre-Luc Bacon , Dacheng Tao

While deep learning is successful in a number of applications, it is not yet well understood theoretically. A satisfactory theoretical characterization of deep learning however, is beginning to emerge. It covers the following questions: 1)…

Machine Learning · Computer Science 2019-08-27 Tomaso Poggio , Andrzej Banburski , Qianli Liao

Dynamical systems theory has recently been applied in optimization to prove that gradient descent algorithms bypass so-called strict saddle points of the loss function. However, in many modern machine learning applications, the required…

Machine Learning · Computer Science 2024-09-12 Patrick Cheridito , Arnulf Jentzen , Florian Rossmannek

This paper extends algorithms that remove the fixed point bias of decentralized gradient descent to solve the more general problem of distributed optimization over subspace constraints. Leveraging the integral quadratic constraint…

Optimization and Control · Mathematics 2022-10-31 Dennis J. Marquis , Dany Abou Jaoude , Mazen Farhood , Craig A. Woolsey

Adversarial training is a popular method to give neural nets robustness against adversarial perturbations. In practice adversarial training leads to low robust training loss. However, a rigorous explanation for why this happens under…

Machine Learning · Computer Science 2020-02-25 Yi Zhang , Orestis Plevrakis , Simon S. Du , Xingguo Li , Zhao Song , Sanjeev Arora