中文
相关论文

相关论文: Existence and Estimation of Critical Batch Size fo…

200 篇论文

Mini-batch stochastic gradient descent and variants thereof have become standard for large-scale empirical risk minimization like the training of neural networks. These methods are usually used with a constant batch size chosen by simple…

机器学习 · 计算机科学 2017-06-29 Lukas Balles , Javier Romero , Philipp Hennig

Generative Adversarial Networks (GANs) are widely used models to learn complex real-world distributions. In GANs, the training of the generator usually stops when the discriminator can no longer distinguish the generator's output from the…

机器学习 · 计算机科学 2021-02-19 Yuanzhi Li , Zehao Dou

The performance of mini-batch stochastic gradient descent (SGD) strongly depends on setting the batch size and learning rate to minimize the empirical loss in training the deep neural network. In this paper, we present theoretical analyses…

机器学习 · 计算机科学 2025-02-17 Hikaru Umeda , Hideaki Iiduka

We present a continual learning approach for generative adversarial networks (GANs), by designing and leveraging parameter-efficient feature map transformations. Our approach is based on learning a set of global and task-specific…

机器学习 · 计算机科学 2021-08-02 Sakshi Varshney , Vinay Kumar Verma , Srijith P K , Lawrence Carin , Piyush Rai

Generative Adversarial Networks (GANs) are a powerful framework for deep generative modeling. Posed as a two-player minimax problem, GANs are typically trained end-to-end on real-valued data and can be used to train a generator of…

机器学习 · 统计学 2017-11-15 Anirudh Goyal , Nan Rosemary Ke , Alex Lamb , R Devon Hjelm , Chris Pal , Joelle Pineau , Yoshua Bengio

Generative Adversarial Networks (GANs) are powerful Machine Learning models capable of generating fully synthetic samples of a desired phenomenon with a high resolution. Despite their success, the training process of a GAN is highly…

机器学习 · 计算机科学 2022-09-07 Ángel González-Prieto , Alberto Mozo , Edgar Talavera , Sandra Gómez-Canaval

Generative Adversarial Networks (GANs) have proven to be a powerful tool for generating realistic synthetic data. However, traditional GANs often struggle to capture complex relationships between features which results in generation of…

机器学习 · 计算机科学 2023-06-06 Srikrishna Iyer , Teng Teck Hou

Gradient temporal difference (Gradient TD) algorithms are a popular class of stochastic approximation (SA) algorithms used for policy evaluation in reinforcement learning. Here, we consider Gradient TD algorithms with an additional heavy…

机器学习 · 计算机科学 2021-11-23 Rohan Deb , Shalabh Bhatnagar

Generative adversarial networks (GANs) are highly effective unsupervised learning frameworks that can generate very sharp data, even for data such as images with complex, highly multimodal distributions. However GANs are known to be very…

机器学习 · 统计学 2017-12-05 Sitao Xiang , Hao Li

Training deep neural networks (DNNs) used in modern machine learning is computationally expensive. Machine learning scientists, therefore, rely on stochastic first-order methods for training, coupled with significant hand-tuning, to obtain…

Generative models trained on finite data face a fundamental tension: their score-matching or next-token objective converges to the empirical training distribution rather than the population distribution we seek to learn. Using rule-valid…

机器学习 · 计算机科学 2026-05-12 Binxu Wang , Emma Lucia Byrnes Finn , Bingbin Liu

The human brain has the ability to carry out new tasks with limited experience. It utilizes prior learning experiences to adapt the solution strategy to new domains. On the other hand, deep neural networks (DNNs) generally need large…

神经与进化计算 · 计算机科学 2022-02-17 Shayan Hassantabar , Prerit Terway , Niraj K. Jha

We study the effect of mini-batching on the loss landscape of deep neural networks using spiked, field-dependent random matrix theory. We demonstrate that the magnitude of the extremal values of the batch Hessian are larger than those of…

机器学习 · 统计学 2021-11-08 Diego Granziol , Stefan Zohren , Stephen Roberts

We investigate the impact of the input dimension on the generalization error in generative adversarial networks (GANs). In particular, we first provide both theoretical and practical evidence to validate the existence of an optimal input…

机器学习 · 计算机科学 2024-05-08 Zhiyao Tan , Ling Zhou , Huazhen Lin

Generative data augmentation, which scales datasets by obtaining fake labeled examples from a trained conditional generative model, boosts classification performance in various learning tasks including (semi-)supervised learning, few-shot…

机器学习 · 计算机科学 2023-05-30 Chenyu Zheng , Guoqiang Wu , Chongxuan Li

Increasing the batch size during training -- a ''batch ramp'' -- is a promising strategy to accelerate large language model pretraining. While for SGD, doubling the batch size can be equivalent to halving the learning rate, the optimal…

机器学习 · 计算机科学 2025-10-17 Alexandru Meterez , Depen Morwani , Jingfeng Wu , Costin-Andrei Oncescu , Cengiz Pehlevan , Sham Kakade

Learning rate, batch size and momentum are three important hyperparameters in the SGD algorithm. It is known from the work of Jastrzebski et al. arXiv:1711.04623 that large batch size training of neural networks yields models which do not…

计算机视觉与模式识别 · 计算机科学 2020-06-23 Sandesh Kamath , Amit Deshpande , K V Subrahmanyam

Generative adversarial networks (GANs) are powerful tools for learning generative models. In practice, the training may suffer from lack of convergence. GANs are commonly viewed as a two-player zero-sum game between two neural networks.…

机器学习 · 计算机科学 2018-07-13 Hao Ge , Yin Xia , Xu Chen , Randall Berry , Ying Wu

We study the role of an essential hyper-parameter that governs the training of Transformers for neural machine translation in a low-resource setting: the batch size. Using theoretical insights and experimental evidence, we argue against the…

计算与语言 · 计算机科学 2022-03-22 Àlex R. Atrio , Andrei Popescu-Belis

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…

机器学习 · 计算机科学 2018-12-18 Sam McCandlish , Jared Kaplan , Dario Amodei , OpenAI Dota Team