This paper presents novel circuit configurations to realize Neuron Activation Function (NAF) and the Derivative of Neuron Activation Function (DNAF) with low power dissipation. The circuits realize NAF and DNAF simultaneously with programmable characteristics. In the proposed configurations, tan sigmoid and log sigmoid NAFs and their derivative can be obtained from the same circuit simultaneous...

This paperpnsenb a radical-partitioned neural network system using a modified Sigmoid fvnction and a weight-dotted radical selector f o r large-volume Chinese characters recognition VLSI. Wiih a modified Sigmoid function and the weight-dotted radical selector, the recognition rate of 1000 radical-partitioned Chinese characters can be enhanced to 90% from 70% for the input samples with 15% rando...

recurrent sigmoid volvulus is a clinical entity characterized by recurrent episodes of partial or complete sigmoid volvulus. although it is commonly seen in the elderly, it can be occasionally seen in younger patients. patients with recurrent partial sigmoid volvulus are relatively asymptomatic or present with mild abdominal pain. early diagnosis and treatment is essential to prevent conversion...

In this paper we propose a new algorithm for learning polyhedral classifiers. In contrast to existing methods for learning polyhedral classifier which solve a constrained optimization problem, our method solves an unconstrained optimization problem. Our method is based on a logistic function based model for the posterior probability function. We propose an alternating optimization algorithm, na...

In this study, we present the operator-dependent sigmoid function, which is derived from a universal unary operator called kappa function. Here, describe how generalized function related to representable uninorms (i.e., Dombi's aggregative operator). Namely, show that inverse of an additive generator operator. We provide necessary and sufficient conditions for form transforms into conjunctive o...

<span lang="EN-US">This study implements a recurrent neural network (RNN) by comparing two RNN structures, namely Elman and Jordan using the backpropagation through time (BPTT) programming algorithm in training forecasting process foreign exchange cases. The activation functions used are linear transfer function, tan-sigmoid function (Tansig), log-sigmoid (Logsig), which applied to hidden...