English
Related papers

Related papers: On the Width Scaling of Neural Optimizers Under Ma…

200 papers

The Muon optimizer has received considerable attention for its strong performance in training large language models, yet the design principle behind its matrix-gradient orthogonalization remains largely elusive. In this paper, we introduce…

Optimization and Control · Mathematics 2026-04-03 Zhehang Du , Weijie Su

The choice of optimizer significantly impacts the training efficiency and computational costs of large language models (LLMs). Recently, the Muon optimizer has demonstrated promising results by orthogonalizing parameter updates, improving…

Machine Learning · Computer Science 2025-10-08 Zichong Li , Liming Liu , Chen Liang , Weizhu Chen , Tuo Zhao

The ability of machine learning (ML) algorithms to generalize well to unseen data has been studied through the lens of information theory, by bounding the generalization error with the input-output mutual information (MI), i.e., the MI…

Machine Learning · Statistics 2024-06-07 Kimia Nadjahi , Kristjan Greenewald , Rickard Brüel Gabrielsson , Justin Solomon

We analyze recurrent neural networks with diagonal hidden-to-hidden weight matrices, trained with gradient descent in the supervised learning setting, and prove that gradient descent can achieve optimality \emph{without} massive…

Machine Learning · Computer Science 2024-10-11 Semih Cayci , Atilla Eryilmaz

This article provides a comprehensive understanding of optimization in deep learning, with a primary focus on the challenges of gradient vanishing and gradient exploding, which normally lead to diminished model representational ability and…

Machine Learning · Computer Science 2023-11-14 Xianbiao Qi , Jianan Wang , Lei Zhang

Margin enlargement over training data has been an important strategy since perceptrons in machine learning for the purpose of boosting the robustness of classifiers toward a good generalization ability. Yet Breiman (1999) showed a dilemma…

Machine Learning · Computer Science 2021-01-05 Weizhi Zhu , Yifei Huang , Yuan Yao

The successful training of deep neural networks requires addressing challenges such as overfitting, numerical instabilities leading to divergence, and increasing variance in the residual stream. A common solution is to apply regularization…

Machine Learning · Computer Science 2025-11-20 Jörg K. H. Franke , Urs Spiegelhalter , Marianna Nezhurina , Jenia Jitsev , Frank Hutter , Michael Hefenbrock

Balancing convergence speed, generalization capability, and computational efficiency remains a core challenge in deep learning optimization. First-order gradient descent methods, epitomized by stochastic gradient descent (SGD) and Adam,…

Recent numerical experiments have demonstrated that the choice of optimization geometry used during training can impact generalization performance when learning expressive nonlinear model classes such as deep neural networks. These…

Machine Learning · Computer Science 2022-04-25 Nicholas M. Boffi , Stephen Tu , Jean-Jacques E. Slotine

Gradient orthogonalization is a simple strategy that shows great utility in speeding up gradient descent. The Muon optimizer (Jordan, Jin, et al., 2024) combines gradient orthogonalization with first-order momentum and achieves significant…

Machine Learning · Computer Science 2025-10-21 Ahmed Khaled , Kaan Ozkara , Tao Yu , Mingyi Hong , Youngsuk Park

We study the implicit bias of momentum-based optimizers on smooth homogeneous models. We show that \textit{momentum steepest descent} algorithms like Muon (spectral norm), MomentumGD ($\ell_2$ norm), and Signum ($\ell_\infty$ norm) are…

Machine Learning · Computer Science 2026-05-26 Eitan Gronich , Gal Vardi

Recently, flat-minima optimizers, which seek to find parameters in low-loss neighborhoods, have been shown to improve a neural network's generalization performance over stochastic and adaptive gradient-based optimizers. Two methods have…

Machine Learning · Computer Science 2023-01-30 Jean Kaddour , Linqing Liu , Ricardo Silva , Matt J. Kusner

Adaptive optimization algorithms, particularly Adam and its variant AdamW, are fundamental components of modern deep learning. However, their training dynamics lack comprehensive theoretical understanding, with limited insight into why…

Machine Learning · Computer Science 2024-12-23 Rhys Gould , Hidenori Tanaka

Sharpness Aware Minimization (SAM) enhances performance across various neural architectures and datasets. As models are continually scaled up to improve performance, a rigorous understanding of SAM's scaling behaviour is paramount. To this…

Machine Learning · Computer Science 2025-02-12 Moritz Haas , Jin Xu , Volkan Cevher , Leena Chennuru Vankadara

Multimodal alignment is commonly learned from isolated image-text pairs via CLIP-style dual encoders, leaving the relational context among entities largely unused. Multimodal attributed graphs (MAGs), where nodes carry multimodal attributes…

Machine Learning · Computer Science 2026-05-18 Xu Wang , Xunkai Li , Yinlin Zhu , Rong-Hua Li , Guoren Wang

This paper investigates the impact of different optimizers on the grokking phenomenon, where models exhibit delayed generalization. We conducted experiments across seven numerical tasks (primarily modular arithmetic) using a modern…

Machine Learning · Computer Science 2025-04-23 Amund Tveit , Bjørn Remseth , Arve Skogvold

Maintaining numerical stability in machine learning models is crucial for their reliability and performance. One approach to maintain stability of a network layer is to integrate the condition number of the weight matrix as a regularizing…

Machine Learning · Computer Science 2024-10-02 Rossen Nenov , Daniel Haider , Peter Balazs

Model merging offers a scalable alternative to multi-task learning but often yields suboptimal performance on classification tasks. We attribute this degradation to a geometric misalignment between the merged encoder and static…

Machine Learning · Computer Science 2026-02-03 Fanshuang Kong , Richong Zhang , Zhijie Nie , Hang Zhou , Ziqiao Wang , Qiang Sun , Chunming Hu

Optimization is an integral part of modern deep learning. Recently, the concept of learned optimizers has emerged as a way to accelerate this optimization process by replacing traditional, hand-crafted algorithms with meta-learned…

Machine Learning · Computer Science 2023-12-13 Jan Sobotka , Petr Šimánek , Daniel Vašata

MLP is a heavily used backbone in modern deep learning (DL) architectures for supervised learning on tabular data, and AdamW is the go-to optimizer used to train tabular DL models. Unlike architecture design, however, the choice of…

Machine Learning · Computer Science 2026-04-20 Yury Gorishniy , Ivan Rubachev , Dmitrii Feoktistov , Artem Babenko