English
Related papers

Related papers: Subsampling is not Magic: Why Large Batch Sizes Wo…

200 papers

Understanding when the noise in stochastic gradient descent (SGD) affects generalization of deep neural networks remains a challenge, complicated by the fact that networks can operate in distinct training regimes. Here we study how the…

Machine Learning · Computer Science 2023-05-31 Antonio Sclocchi , Mario Geiger , Matthieu Wyart

In an increasing number of domains it has been demonstrated that deep learning models can be trained using relatively large batch sizes without sacrificing data efficiency. However the limits of this massive data parallelism seem to differ…

Machine Learning · Computer Science 2018-12-18 Sam McCandlish , Jared Kaplan , Dario Amodei , OpenAI Dota Team

When using large-batch training to speed up stochastic gradient descent, learning rates must adapt to new batch sizes in order to maximize speed-ups and preserve model quality. Re-tuning learning rates is resource intensive, while fixed…

Machine Learning · Computer Science 2020-07-13 Tyler B. Johnson , Pulkit Agrawal , Haijie Gu , Carlos Guestrin

We investigate how the final parameters found by stochastic gradient descent are influenced by over-parameterization. We generate families of models by increasing the number of channels in a base network, and then perform a large…

Machine Learning · Computer Science 2019-05-10 Daniel S. Park , Jascha Sohl-Dickstein , Quoc V. Le , Samuel L. Smith

This paper presents a methodology for selecting the mini-batch size that minimizes Stochastic Gradient Descent (SGD) learning time for single and multiple learner problems. By decoupling algorithmic analysis issues from hardware and…

Machine Learning · Computer Science 2019-11-18 Michael P. Perrone , Haidar Khan , Changhoan Kim , Anastasios Kyrillidis , Jerry Quinn , Valentina Salapura

Differentially private (DP) release of multidimensional statistics typically considers an aggregate sensitivity, e.g. the vector norm of a high-dimensional vector. However, different dimensions of that vector might have widely different…

Machine Learning · Statistics 2022-10-31 Joonas Jälkö , Lukas Prediger , Antti Honkela , Samuel Kaski

Stochastic gradient descent (SGD) and its variants have established themselves as the go-to algorithms for large-scale machine learning problems with independent samples due to their generalization performance and intrinsic computational…

Machine Learning · Statistics 2025-08-25 Hao Chen , Lili Zheng , Raed Al Kontar , Garvesh Raskutti

Optimization objectives in the form of a sum of intractable expectations are rising in importance (e.g., diffusion models, variational autoencoders, and many more), a setting also known as "finite sum with infinite data." For these…

Machine Learning · Statistics 2025-05-13 Kyurae Kim , Joohwan Ko , Yi-An Ma , Jacob R. Gardner

Differentially private stochastic gradient descent (DP-SGD) is broadly considered to be the gold standard for training and fine-tuning neural networks under differential privacy (DP). With the increasing availability of high-quality…

Modern deep networks are trained with stochastic gradient descent (SGD) whose key hyperparameters are the number of data considered at each step or batch size $B$, and the step size or learning rate $\eta$. For small $B$ and large $\eta$,…

Machine Learning · Computer Science 2024-02-29 Antonio Sclocchi , Matthieu Wyart

Differentially Private Stochastic Gradient Descent (DP-SGD) limits the amount of private information deep learning models can memorize during training. This is achieved by clipping and adding noise to the model's gradients, and thus…

Computer Vision and Pattern Recognition · Computer Science 2023-06-22 Florian A. Hölzl , Daniel Rueckert , Georgios Kaissis

To maximize hardware utilization, modern machine learning systems typically employ large constant or manually tuned batch size schedules, relying on heuristics that are brittle and costly to tune. Existing adaptive strategies based on…

It is well known that, for most datasets, the use of large-size minibatches for Stochastic Gradient Descent (SGD) typically leads to slow convergence and poor generalization. On the other hand, large minibatches are of great practical…

Machine Learning · Computer Science 2018-06-20 Matteo Fischetti , Iacopo Mandatelli , Domenico Salvagnin

By ensuring differential privacy in the learning algorithms, one can rigorously mitigate the risk of large models memorizing sensitive training data. In this paper, we study two algorithms for this purpose, i.e., DP-SGD and DP-NSGD, which…

Machine Learning · Computer Science 2022-06-28 Xiaodong Yang , Huishuai Zhang , Wei Chen , Tie-Yan Liu

Large-scale distributed training of deep neural networks results in models with worse generalization performance as a result of the increase in the effective mini-batch size. Previous approaches attempt to address this problem by varying…

Machine Learning · Computer Science 2020-02-17 Kazuki Osawa , Yohei Tsuji , Yuichiro Ueno , Akira Naruse , Chuan-Sheng Foo , Rio Yokota

When applied to large-scale learning problems, the conventional wisdom on privacy-preserving deep learning, known as Differential Private Stochastic Gradient Descent (DP-SGD), has met with limited success due to significant performance…

Machine Learning · Computer Science 2021-12-30 Jian Du , Haitao Mi

We study the learning dynamics of a multi-pass, mini-batch Stochastic Gradient Descent (SGD) procedure for empirical risk minimization in high-dimensional multi-index models with isotropic random data. In an asymptotic regime where the…

Machine Learning · Statistics 2026-02-19 Zhou Fan , Leda Wang

Stochastic gradient descent samples uniformly the training set to build an unbiased gradient estimate with a limited number of samples. However, at a given step of the training process, some data are more helpful than others to continue…

Machine Learning · Computer Science 2023-03-30 Thibault Lahire

We study the role of batch size in stochastic conditional gradient methods under a $\mu$-Kurdyka-{\L}ojasiewicz ($\mu$-KL) condition. Focusing on momentum-based stochastic conditional gradient algorithms (e.g., Scion), we derive a new…

Machine Learning · Computer Science 2026-03-24 Rustem Islamov , Roman Machacek , Aurelien Lucchi , Antonio Silveti-Falls , Eduard Gorbunov , Volkan Cevher

Traditional Differentially Private Stochastic Gradient Descent (DP-SGD) introduces statistical noise on top of gradients drawn from a Gaussian distribution to ensure privacy. This paper introduces the novel Differentially Private Block-wise…

Machine Learning · Computer Science 2025-01-22 David Zagardo