paper-with-me

Papers

Adaptive Dropout with Rademacher Complexity Regularization

2018-01-01 · ICLR 2018 1 · Ke Zhai, Huan Wang

We propose a novel framework to adaptively adjust the dropout rates for the deep neural network based on a Rademacher complexity bound. The state-of-the-art deep learning algorithms impose dropout strategy to prevent feature co-adaptation. However, choosing the dropout rates remains an art of heuristics or relies on empirical grid-search over some hyperparameter space. In this work, we show the network Rademacher complexity is bounded by a function related to the dropout rate vectors and the weight coefficient matrices. Subsequently, we impose this bound as a regularizer and provide a theoretical justified way to trade-off between model complexity and representation power. Therefore, the dropout rates and the empirical loss are unified into the same objective function, which is then optimized using the block coordinate descent algorithm. We discover that the adaptively adjusted dropout rates converge to some interesting distributions that reveal meaningful patterns.Experiments on the task of image and document classification also show our method achieves better performance compared to the state-of the-art dropout algorithms.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Document Classification

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

Dropout Training, Data-dependent Regularization, and Generalization Bounds

2018-07-01 · ICML 2018 7 · Wenlong Mou, Yuchen Zhou, Jun Gao, Li-Wei Wang

We study the problem of generalization guarantees for dropout training. A general framework is first proposed for learning procedures with random perturbation on model parameters. The generalization error is bounded…

Generalization Bounds

Dropout Rademacher Complexity of Deep Neural Networks

2014-02-16 · Wei Gao, Zhi-Hua Zhou

Great successes of deep neural networks have been witnessed in various real applications. Many algorithmic and implementation techniques have been developed, however, theoretical understanding of many aspects of deep neu…

LocalDrop: A Hybrid Regularization for Deep Neural Networks

2021-03-01 · Ziqing Lu, Chang Xu, Bo Du, Takashi Ishida 외

In neural networks, developing regularization algorithms to settle overfitting is one of the major study areas. We propose a new approach for the regularization of neural networks by the local Rademacher complexity calle…

FlexiDrop: Theoretical Insights and Practical Advances in Random Dropout Method on GNNs

2024-05-30 · Zhiheng Zhou, Sihao Liu, Weichen Zhao

Graph Neural Networks (GNNs) are powerful tools for handling graph-type data. Recently, GNNs have been widely applied in various domains, but they also face some issues, such as overfitting, over-smoothing and non-robust…

Effect of Various Regularizers on Model Complexities of Neural Networks in Presence of Input Noise

2019-01-31 · Mayank Sharma, Aayush Yadav, Sumit Soman, Jayadeva

Deep neural networks are over-parameterized, which implies that the number of parameters are much larger than the number of samples used to train the network. Even in such a regime deep architectures do not overfit. This…