paper-with-me

홈 › Papers

SGD and Weight Decay Secretly Minimize the Rank of Your Neural Network

2022-06-12 · Tomer Galanti, Zachary S. Siegel, Aparna Gupte, Tomaso Poggio

We investigate the inherent bias of Stochastic Gradient Descent (SGD) toward learning low-rank weight matrices during the training of deep neural networks. Our results demonstrate that training with mini-batch SGD and weight decay induces a bias toward rank minimization in the weight matrices. Specifically, we show both theoretically and empirically that this bias becomes more pronounced with smaller batch sizes, higher learning rates, or stronger weight decay. Additionally, we predict and empirically confirm that weight decay is essential for this bias to occur. Unlike previous literature, our analysis does not rely on assumptions about the data, convergence, or optimality of the weight matrices, making it applicable to a wide range of neural network architectures of any width or depth. Finally, we empirically explore the connection between this bias and generalization, finding that it has a marginal effect on the test performance.

📄 PDF Abstract BibTeX arXiv:2206.05794

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Weight Decay 설명 없음
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…
Batch Normalization 설명 없음

Similar Papers 제목 키워드 기반

How to decay your learning rate

2021-03-23 · Aitor Lewkowycz

Complex learning rate schedules have become an integral part of deep learning. We find empirically that common fine-tuned schedules decay the learning rate after the weight norm bounces. This leads to the proposal of ABE…

Your Classifier can Secretly Suffice Multi-Source Domain Adaptation

2021-03-20 · NeurIPS 2020 12 · Naveen Venkat, Jogendra Nath Kundu, Durgesh Kumar Singh, Ambareesh Revanur 외

Multi-Source Domain Adaptation (MSDA) deals with the transfer of task knowledge from multiple labeled source domains to an unlabeled target domain, under a domain-shift. Existing methods aim to minimize this domain-shift…

Domain AdaptationMulti-Source Unsupervised Domain Adaptation

Your Data Manifold is Secretly a Reward Model: Shell-LCC for Text-to-Video Generation

2026-06-29 · Shihao Zhang, Yuguang Yan, Junzhe Zhang, Wei Zhao 외 arxiv

Recent text-to-video (T2V) diffusion models rely heavily on auxiliary reward signals (e.g., via reward models or DPO) to align generated content with human aesthetics and improve realism. These signals, however, incur su…

Text-to-Video Generation

Convergence Bound and Critical Batch Size of Muon Optimizer

2025-07-02 · Naoki Sato, Hiroki Naganuma, Hideaki Iiduka arxiv

Muon, a recently proposed optimizer that leverages the inherent matrix structure of neural network parameters, has demonstrated strong empirical performance, indicating its potential as a successor to standard optimizers…

Image Classification

Your Weak LLM is Secretly a Strong Teacher for Alignment

2024-09-13 · Leitian Tao, Yixuan Li

The burgeoning capabilities of large language models (LLMs) have underscored the need for alignment to ensure these models act in accordance with human values and intentions. Existing alignment frameworks present constra…