English
Related papers

Related papers: Fast Convergence of Random Reshuffling under Over-…

200 papers

Random reshuffling with momentum (RRM) corresponds to the SGD optimizer with momentum option enabled, as found in many machine learning libraries like PyTorch and TensorFlow. Despite its widespread use, the convergence properties of RRM do…

Optimization and Control · Mathematics 2026-03-24 Junwen Qiu , Bohao Ma , Andre Milzarek

Random reshuffling, which randomly permutes the dataset each epoch, is widely adopted in model training because it yields faster convergence than with-replacement sampling. Recent studies indicate greedily chosen data orderings can further…

Machine Learning · Computer Science 2023-01-05 Yucheng Lu , Wentao Guo , Christopher De Sa

We prove explicit bounds on the exponential rate of convergence for the momentum stochastic gradient descent scheme (MSGD) for arbitrary, fixed hyperparameters (learning rate, friction parameter) and its continuous-in-time counterpart in…

Optimization and Control · Mathematics 2024-11-07 Benjamin Gess , Sebastian Kassing

We study the random reshuffling (RR) method for smooth nonconvex optimization problems with a finite-sum structure. Though this method is widely utilized in practice such as the training of neural networks, its convergence behavior is only…

Optimization and Control · Mathematics 2023-01-26 Xiao Li , Andre Milzarek , Junwen Qiu

Several works have aimed to explain why overparameterized neural networks generalize well when trained by Stochastic Gradient Descent (SGD). The consensus explanation that has emerged credits the randomized nature of SGD for the bias of the…

Machine Learning · Computer Science 2021-02-24 Shengchao Liu , Dimitris Papailiopoulos , Dimitris Achlioptas

Most modern learning problems are highly overparameterized, meaning that there are many more parameters than the number of training data points, and as a result, the training loss may have infinitely many global minima (parameter vectors…

Machine Learning · Computer Science 2019-06-11 Navid Azizan , Sahin Lale , Babak Hassibi

Most prior work on the convergence of gradient descent (GD) for overparameterized neural networks relies on strong assumptions on the step size (infinitesimal), the hidden-layer width (infinite), or the initialization (large, spectral,…

Machine Learning · Computer Science 2025-05-20 Ziqing Xu , Hancheng Min , Salma Tarmoun , Enrique Mallada , Rene Vidal

Random Reshuffling (RR), also known as Stochastic Gradient Descent (SGD) without replacement, is a popular and theoretically grounded method for finite-sum minimization. We propose two new algorithms: Proximal and Federated Random…

Machine Learning · Computer Science 2021-02-15 Konstantin Mishchenko , Ahmed Khaled , Peter Richtárik

Stochastic gradient descent (SGD) has become the most attractive optimization method in training large-scale deep neural networks due to its simplicity, low computational cost in each updating step, and good performance. Standard excess…

Machine Learning · Computer Science 2021-05-14 Yi Xu , Qi Qian , Hao Li , Rong Jin

We uncover how SGD interacts with batch normalization and can exhibit undesirable training dynamics such as divergence. More precisely, we study how Single Shuffle (SS) and Random Reshuffle (RR) -- two widely used variants of SGD --…

Machine Learning · Computer Science 2023-08-15 David X. Wu , Chulhee Yun , Suvrit Sra

We study stochastic gradient descent {\em without replacement} (\sgdwor) for smooth convex functions. \sgdwor is widely observed to converge faster than true \sgd where each sample is drawn independently {\em with replacement}…

Optimization and Control · Mathematics 2020-02-28 Prateek Jain , Dheeraj Nagaraj , Praneeth Netrapalli

We study the performance of stochastic gradient descent (SGD) on smooth and strongly-convex finite-sum optimization problems. In contrast to the majority of existing theoretical works, which assume that individual functions are sampled with…

Machine Learning · Computer Science 2021-06-03 Itay Safran , Ohad Shamir

Many modern learning tasks involve fitting nonlinear models to data which are trained in an overparameterized regime where the parameters of the model exceed the size of the training dataset. Due to this overparameterization, the training…

Machine Learning · Computer Science 2018-12-27 Samet Oymak , Mahdi Soltanolkotabi

Modern machine learning is dominated by complex, overparameterized architectures capable of interpolating data and achieving zero training loss. For such models, we investigate the convergence properties of two popular modifications to…

Optimization and Control · Mathematics 2026-05-27 Aleksandr Lobanov , Anastasia Koloskova

We analyze the convergence rates of two popular variants of coordinate descent (CD): random CD (RCD), in which the coordinates are sampled uniformly at random, and random-permutation CD (RPCD), in which random permutations are used to…

Optimization and Control · Mathematics 2025-05-30 Donghwa Kim , Jaewook Lee , Chulhee Yun

Accelerated stochastic gradient descent (ASGD) is a workhorse in deep learning and often achieves better generalization performance than SGD. However, existing optimization theory can only explain the faster convergence of ASGD, but cannot…

Machine Learning · Computer Science 2023-11-27 Xuheng Li , Yihe Deng , Jingfeng Wu , Dongruo Zhou , Quanquan Gu

Random reshuffling techniques are prevalent in large-scale applications, such as training neural networks. While the convergence and acceleration effects of random reshuffling-type methods are fairly well understood in the smooth setting,…

Optimization and Control · Mathematics 2025-07-29 Junwen Qiu , Xiao Li , Andre Milzarek

Neural networks have many successful applications, while much less theoretical understanding has been gained. Towards bridging this gap, we study the problem of learning a two-layer overparameterized ReLU neural network for multi-class…

Machine Learning · Computer Science 2019-08-02 Yuanzhi Li , Yingyu Liang

We study stochastic gradient descent (SGD) for composite optimization problems with $N$ sequential operators subject to perturbations in both the forward and backward passes. Unlike classical analyses that treat gradient noise as additive…

Optimization and Control · Mathematics 2026-02-25 Boao Kong , Hengrui Zhang , Kun Yuan

Classical statistical learning theory predicts that overparameterized models should exhibit severe overfitting, yet modern deep neural networks with far more parameters than training samples consistently generalize well. This contradiction…

Machine Learning · Computer Science 2026-04-10 Zeran Johannsen