کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
11030070 | 1646392 | 2019 | 12 صفحه PDF | دانلود رایگان |
عنوان انگلیسی مقاله ISI
Learning structured and non-redundant representations with deep neural networks
ترجمه فارسی عنوان
یادگیری ساختار و بازپرداخت غیر انبوه با شبکه های عصبی عمیق
دانلود مقاله + سفارش ترجمه
دانلود مقاله ISI انگلیسی
رایگان برای ایرانیان
موضوعات مرتبط
مهندسی و علوم پایه
مهندسی کامپیوتر
چشم انداز کامپیوتر و تشخیص الگو
چکیده انگلیسی
This paper proposes a novel regularizer named Structured Decorrelation Constraint, to address both the generalization and optimization of deep neural networks, including multiple-layer perceptrons and convolutional neural networks. Our proposed regularizer reduces overfitting by breaking the co-adaptions between the neurons with an explicit penalty. As a result, the network is capable of learning non-redundant representations. Meanwhile, the proposed regularizer encourages the networks to learn structured high-level features to aid the networks' optimization during training. To this end, neurons are constrained to behave obeying a group prior. Our regularizer applies to various types of layers, including fully connected layers, convolutional layers and normalization layers. The loss of our regularizer can be directly minimized along with the network's classification loss by stochastic gradient descent. Experiments show that the proposed regularizer obviously relieves the overfitting problem of the existing deep networks. It yields much better performance on extensive datasets than the conventional regularizers like Dropout.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Pattern Recognition - Volume 86, February 2019, Pages 224-235
Journal: Pattern Recognition - Volume 86, February 2019, Pages 224-235
نویسندگان
Jihai Yang, Wei Xiong, Shijun Li, Chang Xu,