paper-with-me

Papers

On the SDEs and Scaling Rules for Adaptive Gradient Algorithms

2022-05-20 · Sadhika Malladi, Kaifeng Lyu, Abhishek Panigrahi, Sanjeev Arora

Approximating Stochastic Gradient Descent (SGD) as a Stochastic Differential Equation (SDE) has allowed researchers to enjoy the benefits of studying a continuous optimization trajectory while carefully preserving the stochasticity of SGD. Analogous study of adaptive gradient methods, such as RMSprop and Adam, has been challenging because there were no rigorously proven SDE approximations for these methods. This paper derives the SDE approximations for RMSprop and Adam, giving theoretical guarantees of their correctness as well as experimental validation of their applicability to common large-scaling vision and language settings. A key practical result is the derivation of a $\textit{square root scaling rule}$ to adjust the optimization hyperparameters of RMSprop and Adam when changing batch size, and its empirical validation in deep learning settings.

📄 PDF Abstract BibTeX arXiv:2205.10287

Code (1)

abhishekpanigrahi1996/Adaptive-SDE 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

RMSProp RMSProp is an unpublished adaptive learning rate optimizer proposed by Geoff Hinton. The motivation…
SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Adaptive Methods through the Lens of SDEs: Theoretical Insights on the Role of Noise

2024-11-24 · Enea Monzio Compagnoni, Tianlin Liu, Rustem Islamov, Frank Norbert Proske 외

Despite the vast empirical evidence supporting the efficacy of adaptive optimization methods in deep learning, their theoretical understanding is far from complete. This work introduces novel SDEs for commonly used adapt…

Towards Theoretically Understanding Why SGD Generalizes Better Than ADAM in Deep Learning

2020-10-12 · NeurIPS 2020 12 · Pan Zhou, Jiashi Feng, Chao Ma, Caiming Xiong 외

It is not clear yet why ADAM-alike adaptive gradient algorithms suffer from worse generalization performance than SGD despite their faster training speed. This work aims to provide understandings on this generalization g…

Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs

2025-02-24 · Enea Monzio Compagnoni, Rustem Islamov, Frank Norbert Proske, Aurelien Lucchi

Distributed methods are essential for handling machine learning pipelines comprising large-scale models and datasets. However, their benefits often come at the cost of increased communication overhead between the central…

Distributed OptimizationLanguage ModelingLanguage ModellingQuantization

An SDE for Modeling SAM: Theory and Insights

2023-01-19 · Enea Monzio Compagnoni, Luca Biggio, Antonio Orvieto, Frank Norbert Proske 외

We study the SAM (Sharpness-Aware Minimization) optimizer which has recently attracted a lot of interest due to its increased performance over more classical variants of stochastic gradient descent. Our main contribution…

Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks

2021-05-28 · Dong-Young Lim, Sotirios Sabanis

We present a new class of Langevin based algorithms, which overcomes many of the known shortcomings of popular adaptive optimizers that are currently used for the fine tuning of deep learning models. Its underpinning the…

Deep LearningStochastic Optimization