English
Related papers

Related papers: Directional Consistency as a Complementary Optimiz…

200 papers

Large Language Models (LLMs) are increasingly aligned with human preferences through Reinforcement Learning from Human Feedback (RLHF). Among RLHF methods, Group Relative Policy Optimization (GRPO) has gained attention for its simplicity…

Computation and Language · Computer Science 2025-09-25 Yuhang Zhou , Jing Zhu , Shengyi Qian , Zhuokai Zhao , Xiyao Wang , Xiaoyu Liu , Ming Li , Paiheng Xu , Wei Ai , Furong Huang

Consider the following distributed optimization scenario. A worker has access to training data that it uses to compute the gradients while a server decides when to stop iterative computation based on its target accuracy or delay…

Machine Learning · Computer Science 2022-04-28 Chung-Yi Lin , Victoria Kostina , Babak Hassibi

Out-of-distribution (OOD) generalization poses a serious challenge for modern deep learning (DL). OOD data consists of test data that is significantly different from the model's training data. DL models that perform well on in-domain test…

Computer Vision and Pattern Recognition · Computer Science 2023-08-22 Skylar E. Stolte , Kyle Volle , Aprinda Indahlastari , Alejandro Albizu , Adam J. Woods , Kevin Brink , Matthew Hale , Ruogu Fang

The generalization performance of deep neural networks (DNNs) is a critical factor in achieving robust model behavior on unseen data. Recent studies have highlighted the importance of sharpness-based measures in promoting generalization by…

Machine Learning · Computer Science 2025-01-28 Mohamed Hassan , Aleksandar Vakanski , Boyu Zhang , Min Xian

In this paper, we present enhanced analysis for sign-based optimization algorithms with momentum updates. Traditional sign-based methods, under the separable smoothness assumption, guarantee a convergence rate of $\mathcal{O}(T^{-1/4})$,…

Optimization and Control · Mathematics 2025-07-17 Wei Jiang , Dingzhi Yu , Sifan Yang , Wenhao Yang , Lijun Zhang

Topology optimization by optimally distributing materials in a given domain requires non-gradient optimizers to solve highly complicated problems. However, with hundreds of design variables or more involved, solving such problems would…

Computational Engineering, Finance, and Science · Computer Science 2022-01-27 Changyu Deng , Yizhou Wang , Can Qin , Yun Fu , Wei Lu

To efficiently train large-scale models, low-bit gradient communication compresses full-precision gradients on local GPU nodes into low-precision ones for higher gradient synchronization efficiency among GPU nodes. However, it often…

Machine Learning · Computer Science 2024-12-02 Xingyu Xie , Zhijie Lin , Kim-Chuan Toh , Pan Zhou

Adam is a widely used optimization algorithm in deep learning, yet the specific class of objective functions where it exhibits inherent advantages remains underexplored. Unlike prior studies requiring external schedulers and $\beta_2$ near…

Machine Learning · Computer Science 2026-05-26 Zhiwei Bai , Jiajie Zhao , Zhangchen Zhou , Zhi-Qin John Xu , Yaoyu Zhang

Stochastic gradient descent (SGD) is a standard optimization method to minimize a training error with respect to network parameters in modern neural network learning. However, it typically suffers from proliferation of saddle points in the…

Machine Learning · Computer Science 2017-11-23 Haiping Huang , Taro Toyoizumi

We introduce a new second-order inertial optimization method for machine learning called INNA. It exploits the geometry of the loss function while only requiring stochastic approximations of the function values and the generalized…

Machine Learning · Computer Science 2021-08-17 Camille Castera , Jérôme Bolte , Cédric Févotte , Edouard Pauwels

Many optimizers can be interpreted as steepest-descent methods under norm-induced geometries, and thus inherit corresponding implicit biases. We introduce \nameA{} (\fullname{}), which combines spectral control from orthogonalized update…

Machine Learning · Computer Science 2026-02-10 Zixiao Wang , Yifei Shen , Huishuai Zhang

Despite an extensive body of literature on deep learning optimization, our current understanding of what makes an optimization algorithm effective is fragmented. In particular, we do not understand well whether enhanced optimization…

Machine Learning · Computer Science 2024-03-04 Toki Tahmid Inan , Mingrui Liu , Amarda Shehu

Training instability in modern deep learning systems is frequently triggered by rare but extreme gradient-norm spikes, which can induce oversized parameter updates, corrupt optimizer state, and lead to slow recovery or divergence. Widely…

Deep learning experiments by Cohen et al. [2021] using deterministic Gradient Descent (GD) revealed an Edge of Stability (EoS) phase when learning rate (LR) and sharpness (i.e., the largest eigenvalue of Hessian) no longer behave as in…

Machine Learning · Computer Science 2022-10-31 Sanjeev Arora , Zhiyuan Li , Abhishek Panigrahi

Adaptive gradient-based optimization methods such as \textsc{Adagrad}, \textsc{Rmsprop}, and \textsc{Adam} are widely used in solving large-scale machine learning problems including deep learning. A number of schemes have been proposed in…

Machine Learning · Computer Science 2019-05-30 Parvin Nazari , Davoud Ataee Tarzanagh , George Michailidis

We present consistent optimization for single stage object detection. Previous works of single stage object detectors usually rely on the regular, dense sampled anchors to generate hypothesis for the optimization of the model. Through an…

Computer Vision and Pattern Recognition · Computer Science 2019-01-24 Tao Kong , Fuchun Sun , Huaping Liu , Yuning Jiang , Jianbo Shi

In this study, we dive deep into the inconsistency of pseudo targets in semi-supervised object detection (SSOD). Our core observation is that the oscillating pseudo-targets undermine the training of an accurate detector. It injects noise…

Computer Vision and Pattern Recognition · Computer Science 2023-03-29 Xinjiang Wang , Xingyi Yang , Shilong Zhang , Yijiang Li , Litong Feng , Shijie Fang , Chengqi Lyu , Kai Chen , Wayne Zhang

The rapid development of large language models (LLMs) has driven the demand for more efficient optimization techniques. Among these, the Lookahead family of optimizers employs a two-loop framework, maintaining fast and slow sets of model…

Machine Learning · Computer Science 2025-10-20 Dominik Kallusky , Vinay Rao , Vishal Nandavanam , Hao-Jun Michael Shi

Synchronous federated learning scales poorly due to the straggler effect. Asynchronous algorithms increase the update throughput by processing updates upon arrival, but they introduce two fundamental challenges: gradient staleness, which…

Machine Learning · Computer Science 2026-03-30 Abdelkrim Alahyane , Céline Comte , Matthieu Jonckheere

There has been a recent surge in research on adversarial perturbations that defeat Deep Neural Networks (DNNs) in machine vision; most of these perturbation-based attacks target object classifiers. Inspired by the observation that humans…

Computer Vision and Pattern Recognition · Computer Science 2020-07-27 Shasha Li , Shitong Zhu , Sudipta Paul , Amit Roy-Chowdhury , Chengyu Song , Srikanth Krishnamurthy , Ananthram Swami , Kevin S Chan