English
Related papers

Related papers: Universal Dynamics of Warmup Stable Decay: underst…

200 papers

Adaptive optimization algorithms such as Adam are widely used in deep learning. The stability of such algorithms is often improved with a warmup schedule for the learning rate. Motivated by the difficulty of choosing and tuning warmup…

Machine Learning · Computer Science 2021-03-23 Jerry Ma , Denis Yarats

Training large-scale models presents challenges not only in terms of resource requirements but also in terms of their convergence. For this reason, the learning rate (LR) is often decreased when the size of a model is increased. Such a…

Computation and Language · Computer Science 2025-05-30 Marco Gaido , Sara Papi , Luisa Bentivogli , Alessio Brutti , Mauro Cettolo , Roberto Gretter , Marco Matassoni , Mohamed Nabih , Matteo Negri

Adaptive optimizers such as Adam (Kingma & Ba, 2015) have been central to the success of large language models. However, they often require to maintain optimizer states throughout training, which can result in memory requirements several…

Machine Learning · Computer Science 2025-02-24 Chao Ma , Wenbo Gong , Meyer Scetbon , Edward Meeds

Layer-wise preconditioning methods are a family of memory-efficient optimization algorithms that introduce preconditioners per axis of each layer's weight tensors. These methods have seen a recent resurgence, demonstrating impressive…

Machine Learning · Computer Science 2025-02-05 Thomas T. Zhang , Behrad Moniri , Ansh Nagwekar , Faraz Rahman , Anton Xue , Hamed Hassani , Nikolai Matni

We study adaptive learning rate scheduling for norm-constrained optimizers (e.g., Muon and Lion). We introduce a generalized smoothness assumption under which local curvature decreases with the suboptimality gap and empirically verify that…

Machine Learning · Computer Science 2026-05-19 Artem Riabinin , Andrey Veprikov , Arman Bolatov , Martin Takáč , Aleksandr Beznosikov

Learning rate scheduling plays a critical role in the optimization of deep neural networks, directly influencing convergence speed, stability, and generalization. While existing schedulers such as cosine annealing, cyclical learning rates,…

Machine Learning · Computer Science 2026-03-05 Ayush K. Varshney , Šarūnas Girdzijauskas , Konstantinos Vandikas , Aneta Vulgarakis Feljan

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

We investigate the convergence rates and data sample sizes required for training a machine learning model using a stochastic gradient descent (SGD) algorithm, where data points are sampled based on either their loss value or uncertainty…

Machine Learning · Computer Science 2024-11-26 Daniel Haimovich , Dima Karamshuk , Fridolin Linder , Niek Tax , Milan Vojnovic

Despite the non-convex optimization landscape, over-parametrized shallow networks are able to achieve global convergence under gradient descent. The picture can be radically different for narrow networks, which tend to get stuck in…

Machine Learning · Statistics 2023-06-16 Rodrigo Veiga , Ludovic Stephan , Bruno Loureiro , Florent Krzakala , Lenka Zdeborová

Stochastic Gradient Descent (SGD) stands as a cornerstone optimization algorithm with proven real-world empirical successes but relatively limited theoretical understanding. Recent research has illuminated a key factor contributing to its…

Machine Learning · Computer Science 2024-01-24 Gregory Dexter , Borja Ocejo , Sathiya Keerthi , Aman Gupta , Ayan Acharya , Rajiv Khanna

Commonly used optimization algorithms often show a trade-off between good generalization and fast training times. For instance, stochastic gradient descent (SGD) tends to have good generalization; however, adaptive gradient methods have…

Machine Learning · Computer Science 2023-06-14 Aditya Cowsik , Tankut Can , Paolo Glorioso

We show that learning-rate schedules for large model training behave surprisingly similar to a performance bound from non-smooth convex optimization theory. We provide a bound for the constant schedule with linear cooldown; in particular,…

Machine Learning · Computer Science 2025-07-24 Fabian Schaipp , Alexander Hägele , Adrien Taylor , Umut Simsekli , Francis Bach

The learning rate is a crucial hyperparameter in deep learning, with its ideal value depending on the problem and potentially changing during training. In this paper, we investigate the practical utility of adaptive learning rate mechanisms…

Machine Learning · Computer Science 2025-08-08 Asma Atamna , Tom Maus , Fabian Kievelitz , Tobias Glasmachers

Modern machine learning often requires training with large batch size, distributed data, and massively parallel compute hardware (like mobile and other edge devices or distributed data centers). Communication becomes a major bottleneck in…

Machine Learning · Computer Science 2025-12-12 Ahmed Khaled , Satyen Kale , Arthur Douillard , Chi Jin , Rob Fergus , Manzil Zaheer

It is widely believed that stochastic gradient descent (SGD) performs significantly worse than adaptive optimizers such as Adam in pre-training Large Language Models (LLMs). Yet the underlying reason for this gap remains unclear. In this…

Machine Learning · Computer Science 2026-05-19 Athanasios Glentis , Dawei Li , Chung-Yiu Yau , Mingyi Hong

A basic unanswered question in neural network training is: what is the best learning rate schedule shape for a given workload? The choice of learning rate schedule is a key factor in the success or failure of the training process, but…

Machine Learning · Computer Science 2026-03-16 Hiroki Naganuma , Atish Agarwala , Priya Kasimbeg , George E. Dahl

Optimal Transport has sparked vivid interest in recent years, in particular thanks to the Wasserstein distance, which provides a geometrically sensible and intuitive way of comparing probability measures. For computational reasons, the…

Machine Learning · Computer Science 2024-03-19 Eloi Tanguy

Applying weight decay (WD) to matrix layers is standard practice in large-language-model pretraining. Prior work suggests that stochastic gradient noise induces a Brownian-like expansion of the weight matrices W, whose growth is…

Machine Learning · Computer Science 2026-01-09 Maksim Velikanov , Ilyas Chahed , Jingwei Zuo , Dhia Eddine Rhaiem , Younes Belkada , Hakim Hacid

The success of deep learning can be attributed to various factors such as increase in computational power, large datasets, deep convolutional neural networks, optimizers etc. Particularly, the choice of optimizer affects the generalization,…

Machine Learning · Computer Science 2021-09-10 Anirudh Maiya , Inumella Sricharan , Anshuman Pandey , Srinivas K. S

Stochastic gradient descent (SGD) is a widely used algorithm in machine learning, particularly for neural network training. Recent studies on SGD for canonical quadratic optimization or linear regression show it attains well generalization…

Machine Learning · Computer Science 2024-09-17 Haihan Zhang , Yuanshi Liu , Qianwen Chen , Cong Fang