Integer-Weight Approximation of Continuous-Weight Multilayer Feedforward Nets
نویسندگان
چکیده
منابع مشابه
Weight Perturbation: An Optimal Architecture and Learning Technique for Analog VLSI Feedforward and Recurrent Multilayer Networks
Previous work on analog VLSI implementation of multilayer perceptrons with on-chip learning has mainly targeted the implementation of algorithms such as back-propagation. Although back-propagation is efficient, its implementation in analog VLSI requires excessive computational hardware. It is shown that using gradient descent with direct approximation of the gradient instead of back-propagation...
متن کاملA provably good multilayer topological planar routing algorithm in IC layout designs
Given a number of routing layers, the multilayer topological planar routing problem is to choose a maximum (weighted) set of nets so that each net in the set can be topologically routed entirely in one of the given layers without crossing other nets. This problem has important application in the layout design of multilayer IC technology, which has become available recently. In this paper, we pr...
متن کاملOn the approximation by single hidden layer feedforward neural networks with fixed weights
Single hidden layer feedforward neural networks (SLFNs) with fixed weights possess the universal approximation property provided that approximated functions are univariate. But this phenomenon does not lay any restrictions on the number of neurons in the hidden layer. The more this number, the more the probability of the considered network to give precise results. In this note, we constructivel...
متن کاملThe Method of Steepest Descent for Feedforward Artificial Neural Networks
In this paper, we implement the method of Steepest Descent in single and multilayer feedforward artificial neural networks. In all previous works, all the update weight equations for single or multilayer feedforward artificial neural networks has been calculated by choosing a single activation function for various processing unit in the network. We, at first, calculate the total error function ...
متن کاملMultilayer feedforward networks with a nonpolynomial activation function can approximate any function
-Several researchers characterized the activation fimction under which multilayer feedforward networks can act as universal approximators. We show that most o f all the characterizations that were reported thus far in the literature are special cases o f the following general result: A standard multilayer feedforward network with a locally bounded piecewise continuous activation fimction can ap...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1996