paper-with-me

Papers

Do deep nets really need weight decay and dropout?

2018-02-20 · Alex Hernández-García, Peter König

The impressive success of modern deep neural networks on computer vision tasks has been achieved through models of very large capacity compared to the number of available training examples. This overparameterization is often said to be controlled with the help of different regularization techniques, mainly weight decay and dropout. However, since these techniques reduce the effective capacity of the model, typically even deeper and wider architectures are required to compensate for the reduced capacity. Therefore, there seems to be a waste of capacity in this practice. In this paper we build upon recent research that suggests that explicit regularization may not be as important as widely believed and carry out an ablation study that concludes that weight decay and dropout may not be necessary for object recognition if enough data augmentation is introduced.

📄 PDF Abstract BibTeX arXiv:1802.07042

Code (1)

oliviawl/image_classification_utkface tf

Tasks

Data AugmentationObject Recognition

Methods 이 논문이 사용한 방법론

Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

Surprising properties of dropout in deep networks

2016-02-14 · David P. Helmbold, Philip M. Long

We analyze dropout in deep networks with rectified linear units and the quadratic loss. Our results expose surprising differences between the behavior of dropout and more traditional regularizers like weight decay. For e…

Data augmentation instead of explicit regularization

2018-06-11 · ICLR 2018 1 · Alex Hernández-García, Peter König

Contrary to most machine learning models, modern deep artificial neural networks typically include multiple components that contribute to regularization. Despite the fact that some (explicit) regularization techniques, s…

Data AugmentationObject Categorization

Variational Nested Dropout

2021-01-27 · CVPR 2021 1 · Yufei Cui, Yu Mao, Ziquan Liu, Qiao Li 외

Nested dropout is a variant of dropout operation that is able to order network parameters or features based on the pre-defined importance during training. It has been explored for: I. Constructing nested nets: the nested…

Representation Learning

Do Deep Convolutional Nets Really Need to be Deep and Convolutional?

2016-03-17 · Gregor Urban, Krzysztof J. Geras, Samira Ebrahimi Kahou, Ozlem Aslan 외

Yes, they do. This paper provides the first empirical demonstration that deep convolutional models really need to be both deep and convolutional, even when trained with methods such as distillation that allow small or sh…

Why Do We Need Weight Decay in Modern Deep Learning?

2023-10-06 · Francesco D'Angelo, Maksym Andriushchenko, Aditya Varre, Nicolas Flammarion

Weight decay is a broadly used technique for training state-of-the-art deep networks from image classification to large language models. Despite its widespread usage and being extensively studied in the classical literat…

Deep Learningimage-classificationImage ClassificationLearning Theory+1