Local Coupled Extreme Learning Machine (LCELM) is a recently-proposed variant of ELM, which assigns an address for each hidden-layer node and activates the hidden-layer node when its activated degree is less than a given threshold. In this paper, an improved version of LCELM is proposed by developing a new way to initialize the address for each hidden-layer node and calculating the activated de...

To infer multilayer deep representations of high-dimensional discrete and nonnegative real vectors, we propose an augmentable gamma belief network (GBN) that factorizes each of its hidden layers into the product of a sparse connection weight matrix and the nonnegative real hidden units of the next layer. The GBN’s hidden layers are jointly trained with an upward-downward Gibbs sampler that solv...

A novel algorithm for weight adjustments in a multilayer neural network is derived using the principles of dynamic programming. The algorithm computes the optimal values for weights on a layer-by-layer basis starting from the output layer of the network. The advantage of this algorithm is that it provides an error function for every hidden layer expressed entirely in terms of the weights and ou...

in this study, response surface methodology (rsm) was used to optimize osmo-dehydration of orange slice. effect of osmotic solution temperature in the range of 30 to 60 °c, immersion time from 0 to 300 min and sucrose concentration from 35 to 65 brix degree on water loss, solid gain, moisture content, water loss to solid gain ratio and brix change were investigated by central composite design (...

Prediction of the gas holdup and pressure drop in a horizontal pipe for gas-non-Newtonian liquid flow using Artificial Neural Networks (ANN) methodology have been reported in this paper from the data acquired from our earlier experiment. The ANN prediction is done using Multilayer Perceptrons (MLP) trained with three different algorithms, namely: Backpropagation (BP), Scaled Conjugate gradient ...

We study neural network models that learn location invariant orthographic representations for printed words. We compare two model architectures: with and without a hidden layer. We find that both architectures succeed in learning the training data and in capturing benchmark phenomena of skilled reading – transposed-letter and relative-position priming. Networks without a hidden layer use a stra...

This paper proposes an architecture for deep neural networks with hidden layer branches that learn targets of lower hierarchy than final layer targets. The branches provide a channel for enforcing useful information in hidden layer which helps in attaining better accuracy, both for the final layer and hidden layers. The shared layers modify their weights using the gradients of all cost function...

the use of neural networks methodology is not as common in the investigation and pre-diction noise as statistical analysis. the application of artificial neural networks for pre-diction of power tiller noise is set out in the present paper. the sound pressure signals for noise analysis were obtained in a field experiment using a 13-hp power tiller. during measurement and recording of the sound ...

A total of 1099 data points consisting of alcohol-alcohol, alcohol-alkane, alkane-alkane, alcohol-amine and acid-acid binary solutions were collected from scientific literature to develop an appropriate artificial neural network (ANN) model. Temperature, molecular weight of the pure components, mole fraction of one component and the structural groups of the components were used as input paramet...

We study the number of hidden layers required by a multilayer neu-ral network with threshold units to compute a function f from n d to {O, I}. In dimension d = 2, Gibson characterized the functions computable with just one hidden layer, under the assumption that there is no "multiple intersection point" and that f is only defined on a compact set. We consider the restriction of f to the neighbo...