|
|
المؤلف |
Godfrey,Luke B. (Author)
Gashler,Michael S. (Author)
|
عنوان المجلة |
IEEE Transactions on Neural Networks and Learning Systems
|
المجلد / العدد |
2018/JUL V.29 N.7
|
الوصف المادي |
p 2973 - 2985
|
الموضوع الرئيسي |
الحاسوب
|
الخلاصة |
We present a neural network technique for the analysis and extrapolation of time-series data called neural decomposition (ND). Units with a sinusoidal activation function are used to perform a Fourier-like decomposition of training samples into a sum of sinusoids, augmented by units with nonperiodic activation functions to capture linear trends and other nonperiodic components. We show how careful weight initialization can be combined with regularization to form a simple model that generalizes well.
المزيد
|
رؤوس الموضوعات |
Neural networks (Computer science)
|
رقم التصنيف |
006 ITN
|
موقع المقال |
قاعة الدوريات - الطابق الثاني
|