Augmenting Recurrent Neural Networks Resilience by Dropout

首页> 外文期刊>Neural Networks and Learning Systems, IEEE Transactions on >Augmenting Recurrent Neural Networks Resilience by Dropout

【24h】

Augmenting Recurrent Neural Networks Resilience by Dropout

机译：通过辍学增强递归神经网络的弹性

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

This brief discusses the simple idea that dropout regularization can be used to efficiently induce resiliency to missing inputs at prediction time in a generic neural network. We show how the approach can be effective on tasks where imputation strategies often fail, namely, involving recurrent neural networks and scenarios where whole sequences of input observations are missing. The experimental analysis provides an assessment of the accuracy-resiliency tradeoff in multiple recurrent models, including reservoir computing methods, and comprising real-world ambient intelligence and biomedical time series.

机译：本文简要讨论了一个简单的想法，即可以使用掉落正则化来在通用神经网络中的预测时有效地诱导对丢失输入的弹性。我们展示了该方法如何在插补策略经常失败的任务上有效，即涉及递归神经网络和缺少输入观察的整个序列的场景。实验分析提供了对多个循环模型（包括储层计算方法）的准确性-弹性权衡的评估，包括真实世界的环境情报和生物医学时间序列。

著录项

来源
《Neural Networks and Learning Systems, IEEE Transactions on》 |2020年第1期|345-351|共7页
作者

展开▼
作者单位

Univ Pisa Dipartimento Informat I-56127 Pisa Italy;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词
Training; Biological neural networks; Neurons; Recurrent neural networks; Resilience; Data models; Computational modeling; Deep learning; dependable machine learning; dropout; missing inputs; recurrent neural networks (RNNs);

机译：训练;生物神经网络;神经元;递归神经网络;弹性;数据模型;计算建模;深度学习;可靠的机器学习;退出;缺少输入;递归神经网络（RNN）;

相似文献

外文文献
中文文献
专利

1. FARM: A Flexible Accelerator for Recurrent and Memory Augmented Neural Networks [J] . Challapalle Nagadastagiri, Rampalli Sahithi, Jao Nicholas, Journal of VLSI signal processing systems for signal, image, and video technology . 2020,第11期

机译：农场：用于复发和内存增强神经网络的灵活加速器
2. A Normalized Adaptive Training of Recurrent Neural Networks With Augmented Error Gradient [J] . Yilei W., Qing S., Sheng L. IEEE Transactions on Neural Networks . 2008,第2期

机译：具有增加误差梯度的递归神经网络的归一化自适应训练
3. An augmented CRTRL for complex-valued recurrent neural networks. [J] . Goh SL, Mandic DP Neural Networks: The Official Journal of the International Neural Network Society . 2007,第10期

机译：复数值递归神经网络的增强CRTRL。
4. Adversarial Dropout for Recurrent Neural Networks [C] . Sungrae Park, Kyungwoo Song, Mingi Ji, AAAI Conference on Artificial Intelligence . 2019

机译：经常性神经网络的对抗辍学
5. Gene expression temporal patterns classification with hierarchical Bayesian neural networks and time lagged recurrent neural networks. [D] . Liang, Yulan. 2003

机译：利用分层贝叶斯神经网络和时滞递归神经网络对基因表达时间模式进行分类。
6. Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy [O] . He‐Ming Huang, Yu Xiao, Rui Yang, 2020

机译：基于高分类精度的神经网络中随机忆故障的辍学神经元单位的实现
7. Adversarial Dropout for Recurrent Neural Networks [O] . Sungrae Park, Kyungwoo Song, Mingi Ji, 2019

机译：经常性神经网络的对抗辍学

Augmenting Recurrent Neural Networks Resilience by Dropout

摘要

著录项

相似文献

相关主题

期刊订阅