English
Related papers

Related papers: AdaGC: Improving Training Stability for Large Lang…

200 papers

Adaptive gradient methods like Adagrad and its variants are widespread in large-scale optimization. However, their use of diagonal preconditioning matrices limits the ability to capture parameter correlations. Full-matrix adaptive methods,…

Machine Learning · Computer Science 2025-09-01 Tatyana Matveeva , Aleksandr Katrutsa , Evgeny Frolov

Training large language models is an expensive, compute-bound process that must be repeated as models scale, algorithms improve, and new data is collected. To address this, next-generation hardware accelerators increasingly support…

Machine Learning · Computer Science 2025-06-27 Huangyuan Su , Mujin Kwun , Stephanie Gil , Sham Kakade , Nikhil Anand

Prompt engineering is crucial for fully leveraging large language models (LLMs), yet most existing optimization methods follow a single trajectory, resulting in limited adaptability, gradient conflicts, and high computational overhead. We…

Artificial Intelligence · Computer Science 2026-02-04 Yichen Han , Yuhang Han , Siteng Huang , Guanyu Liu , Zhengpeng Zhou , Bojun Liu , Yujia Zhang , Isaac N Shi , Lewei He , Tianyu Shi

Vision-language models like CLIP have demonstrated remarkable zero-shot transfer capabilities. However, their susceptibility to imperceptible adversarial perturbations remains a critical security concern. While test-time defenses offer a…

Computer Vision and Pattern Recognition · Computer Science 2026-05-18 Zhiwei Li , Jiacheng Xue , Weining Wang , Ajian Liu , Xingyu Gao , Zhenan Sun , Qi Li

Adaptive gradient methods are typically used for training over-parameterized models. To better understand their behaviour, we study a simplistic setting -- smooth, convex losses with models over-parameterized enough to interpolate the data.…

Machine Learning · Computer Science 2021-02-22 Sharan Vaswani , Issam Laradji , Frederik Kunstner , Si Yi Meng , Mark Schmidt , Simon Lacoste-Julien

This paper examines the phenomenon of probabilistic robustness overestimation in TRADES, a prominent adversarial training method. Our study reveals that TRADES sometimes yields disproportionately high PGD validation accuracy compared to the…

Machine Learning · Computer Science 2024-10-11 Jonathan Weiping Li , Ren-Wei Liang , Cheng-Han Yeh , Cheng-Chang Tsai , Kuanchun Yu , Chun-Shien Lu , Shang-Tse Chen

Hyperparameter tuning, particularly the selection of an appropriate learning rate in adaptive gradient training methods, remains a challenge. To tackle this challenge, in this paper, we propose a novel parameter-free optimizer,…

Machine Learning · Computer Science 2024-05-28 Yijiang Pang , Shuyang Yu , Bao Hoang , Jiayu Zhou

Masked diffusion models (MDMs) are a promising alternative to autoregressive models (ARMs), but they suffer from inherently much higher training variance. High variance leads to noisier gradient estimates and unstable optimization, so even…

Machine Learning · Computer Science 2026-05-22 Mengni Jia , Mengyu Zhou , Yihao Liu , Xiaoxi Jiang , Guanjun Jiang

Learning with guidance has proven effective across a wide range of machine learning systems. Guidance may, for example, come from annotated datasets in supervised learning, pseudo-labels in semi-supervised learning, and expert demonstration…

Machine Learning · Computer Science 2025-10-01 Rui Liu , Peng Gao , Yu Shen , Ming Lin , Pratap Tokekar

Stochastic gradient descent (SGD) is one of the most widely used optimization methods for solving various machine learning problems. SGD solves an optimization problem by iteratively sampling a few data points from the input data, computing…

Machine Learning · Computer Science 2020-11-18 Aditya Devarakonda , James Demmel

Adaptive gradient methods have become popular in optimizing deep neural networks; recent examples include AdaGrad and Adam. Although Adam usually converges faster, variations of Adam, for instance, the AdaBelief algorithm, have been…

Machine Learning · Computer Science 2024-10-29 Kushal Chakrabarti , Nikhil Chopra

Deep learning networks are typically trained by Stochastic Gradient Descent (SGD) methods that iteratively improve the model parameters by estimating a gradient on a very small fraction of the training data. A major roadblock faced when…

Machine Learning · Computer Science 2020-06-11 Tao Lin , Lingjing Kong , Sebastian U. Stich , Martin Jaggi

Deep neural network training involves both forward propagation (from features through logits to loss) and backward propagation (from loss through gradients to parameter updates). While perturbations along the forward chain, including…

Machine Learning · Computer Science 2026-05-29 Hua Li

Gradient clipping is a standard training technique used in deep learning applications such as large-scale language modeling to mitigate exploding gradients. Recent experimental studies have demonstrated a fairly special behavior in the…

Machine Learning · Computer Science 2023-06-06 Amirhossein Reisizadeh , Haochuan Li , Subhro Das , Ali Jadbabaie

This work focuses on the decentralized deep learning optimization framework. We propose Adjacent Leader Decentralized Gradient Descent (AL-DSGD), for improving final model performance, accelerating convergence, and reducing the…

Machine Learning · Computer Science 2024-08-21 Haoze He , Jing Wang , Anna Choromanska

Adaptive gradient methods such as AdaGrad and its variants update the stepsize in stochastic gradient descent on the fly according to the gradients received along the way; such methods have gained widespread use in large-scale optimization…

Machine Learning · Statistics 2021-04-20 Rachel Ward , Xiaoxia Wu , Leon Bottou

Although stochastic gradient descent (SGD) method and its variants (e.g., stochastic momentum methods, AdaGrad) are the choice of algorithms for solving non-convex problems (especially deep learning), there still remain big gaps between the…

Optimization and Control · Mathematics 2019-03-07 Zaiyi Chen , Zhuoning Yuan , Jinfeng Yi , Bowen Zhou , Enhong Chen , Tianbao Yang

Asynchronous Federated Learning (AFL) confronts inherent challenges arising from the heterogeneity of devices (e.g., their computation capacities) and low-bandwidth environments, both potentially causing stale model updates (e.g., local…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-07-09 Jiajun Song , Jiajun Luo , Rongwei Lu , Shuzhao Xie , Bin Chen , Zhi Wang

Early stopping monitors global validation loss and halts all parameter updates simultaneously, which is computationally costly for large transformers due to the extended time required for validation inference. We propose \textit{GradES}, a…

Machine Learning · Computer Science 2025-10-20 Qifu Wen , Xi Zeng , Zihan Zhou , Shuaijun Liu , Mehdi Hosseinzadeh , Ningxin Su , Reza Rawassizadeh

Differential privacy is widely employed in decentralized learning to safeguard sensitive data by introducing noise into model updates. However, existing approaches that use fixed-variance noise often degrade model performance and reduce…

Machine Learning · Computer Science 2025-10-24 Xiaoming Wu , Teng Liu , Xin Wang , Ming Yang , Jiguo Yu