English
Related papers

Related papers: On the Interplay Between Stepsize Tuning and Progr…

200 papers

Neural networks (NNs) are central to modern machine learning and achieve state-of-the-art results in many applications. However, the relationship between loss geometry and generalization is still not well understood. The local geometry of…

Machine Learning · Computer Science 2026-04-15 Yuto Omae , Kazuki Sakai , Yohei Kakimoto , Makoto Sasaki , Yusuke Sakai , Hirotaka Takahashi

We introduce a technique for tuning the learning rate scale factor of any base optimization algorithm and schedule automatically, which we call \textsc{mechanic}. Our method provides a practical realization of recent theoretical reductions…

Machine Learning · Computer Science 2023-06-05 Ashok Cutkosky , Aaron Defazio , Harsh Mehta

Stochastic gradient descent is the method of choice for large scale optimization of machine learning objective functions. Yet, its performance is greatly variable and heavily depends on the choice of the stepsizes. This has motivated a…

Machine Learning · Statistics 2019-02-28 Xiaoyu Li , Francesco Orabona

Loss spikes commonly emerge during neural network training with the Adam optimizer across diverse architectures and scales, yet their underlying mechanism remains elusive. While previous explanations attribute these phenomena to sharper…

Machine Learning · Computer Science 2026-05-26 Zhiwei Bai , Zhangchen Zhou , Jiajie Zhao , Xiaolong Li , Zhiyu Li , Feiyu Xiong , Hongkang Yang , Yaoyu Zhang , Zhi-Qin John Xu

Supervised and preference-based fine-tuning techniques have become popular for aligning large language models (LLMs) with user intent and correctness criteria. However, real-world training data often exhibits spurious correlations --…

Computation and Language · Computer Science 2025-05-12 Julia Shuieh , Prasann Singhal , Apaar Shanker , John Heyer , George Pu , Samuel Denton

Recently, the stochastic Polyak step size (SPS) has emerged as a competitive adaptive step size scheme for stochastic gradient descent. Here we develop ProxSPS, a proximal variant of SPS that can handle regularization terms. Developing a…

Optimization and Control · Mathematics 2023-05-05 Fabian Schaipp , Robert M. Gower , Michael Ulbrich

The performance of deep (reinforcement) learning systems crucially depends on the choice of hyperparameters. Their tuning is notoriously expensive, typically requiring an iterative training process to run for numerous steps to convergence.…

Machine Learning · Computer Science 2021-01-19 Vu Nguyen , Sebastian Schulze , Michael A Osborne

Large batch size training of Neural Networks has been shown to incur accuracy loss when trained with the current methods. The exact underlying reasons for this are still not completely understood. Here, we study large batch size training…

Computer Vision and Pattern Recognition · Computer Science 2021-04-21 Zhewei Yao , Amir Gholami , Qi Lei , Kurt Keutzer , Michael W. Mahoney

Large pretrained models can be privately fine-tuned to achieve performance approaching that of non-private models. A common theme in these results is the surprising observation that high-dimensional models can achieve favorable…

Personalisation is a standard feature of conversational AI systems used by millions; yet, the efficacy of personalisation methods is often evaluated in academic research using simulated users rather than real people. This raises questions…

Computation and Language · Computer Science 2026-05-14 Hannah Rose Kirk , Liu Leqi , Fanzhi Zeng , Henry Davidson , Bertie Vidgen , Christopher Summerfield , Scott A. Hale

The Sharpness Aware Minimization (SAM) optimization algorithm has been shown to control large eigenvalues of the loss Hessian and provide generalization benefits in a variety of settings. The original motivation for SAM was a modified loss…

Machine Learning · Computer Science 2023-02-20 Atish Agarwala , Yann N. Dauphin

Recent works have shown that line search methods greatly increase performance of traditional stochastic gradient descent methods on a variety of datasets and architectures [1], [2]. In this work we succeed in extending line search methods…

Machine Learning · Computer Science 2024-03-28 Philip Kenneweg , Leonardo Galli , Tristan Kenneweg , Barbara Hammer

When training neural networks with full-batch gradient descent (GD) and step size $\eta$, the largest eigenvalue of the Hessian -- the sharpness $S(\boldsymbol{\theta})$ -- rises to $2/\eta$ and hovers there, a phenomenon termed the Edge of…

Machine Learning · Computer Science 2026-04-24 Fangshuo Liao , Afroditi Kolomvaki , Anastasios Kyrillidis

We study the implicit bias of the general family of steepest descent algorithms with infinitesimal learning rate in deep homogeneous neural networks. We show that: (a) an algorithm-dependent geometric margin starts increasing once the…

Machine Learning · Computer Science 2025-09-23 Nikolaos Tsilivis , Eitan Gronich , Julia Kempe , Gal Vardi

The overall performance or expected excess risk of an iterative machine learning algorithm can be decomposed into training error and generalization error. While the former is controlled by its convergence analysis, the latter can be tightly…

Machine Learning · Statistics 2018-04-06 Yuansi Chen , Chi Jin , Bin Yu

Empirically, modern deep learning training often occurs at the Edge of Stability (EoS), where the sharpness of the loss exceeds the threshold below which classical convergence analysis applies. Despite recent progress, existing theoretical…

Machine Learning · Computer Science 2026-04-14 Eric Gan

We study the performance of stochastic first-order methods for finding saddle points of convex-concave functions. A notorious challenge faced by such methods is that the gradients can grow arbitrarily large during optimization, which may…

Machine Learning · Computer Science 2024-06-10 Gergely Neu , Nneka Okolo

We study when the \emph{optimization curve} of first-order methods -- the sequence \${f(x\_n)}*{n\ge0}\$ produced by constant-stepsize iterations -- is convex, equivalently when the forward differences \$f(x\_n)-f(x*{n+1})\$ are…

Optimization and Control · Mathematics 2025-09-12 Le Duc Hieu

In this paper, we suggest a new framework for analyzing primal subgradient methods for nonsmooth convex optimization problems. We show that the classical step-size rules, based on normalization of subgradient, or on the knowledge of optimal…

Optimization and Control · Mathematics 2023-11-27 Yurii Nesterov

We study the statistical and computational complexities of the Polyak step size gradient descent algorithm under generalized smoothness and Lojasiewicz conditions of the population loss function, namely, the limit of the empirical loss…

Machine Learning · Computer Science 2021-10-18 Tongzheng Ren , Fuheng Cui , Alexia Atsidakou , Sujay Sanghavi , Nhat Ho