A stochastic backpropagation algorithm for training neural networks

Chen, Y.Q.; Yin, Tang-Kai; Babri, Haroon A.

doi:10.1109/icics.1997.652068

Cited by 7 publications

(5 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The standard approach of feature generation G(n) of specific iteration (i) is based on a series of virtual binary image generation. It is calculated as the sum of output quantities (Yi) of activated neurons in the given iteration step [19,20]:…”

Section: B Pcnn and Feature Generationmentioning

confidence: 99%

See 1 more Smart Citation

A Proposed Model for Standard Arabic Sign Language Recognition Based on Multiplicative Neural Network

Samir

Aboulela

Tolba

2014

The Egyptian Journal of Language Engineering

View full text Add to dashboard Cite

Sign language recognition is one of the most challenging fields in Human-Computer Interface (HCI) applications. Although there are many obstacles that could dramatically limit the spread of sign language translators in our daily life, the community needs for these translators are no longer a luxury and increase day after day, other than the problems of sign languages all over the world, Arabic sign language enjoys its own difficulties and issues. This paper discusses Arabic sign language problems and proposes a recognition model for standard Arabic sign language. A model is proposed and developed for real-time hand signs recognition. The experiment was conducted on 100signs and the result was 94% recognition accuracy confirming words offline extendibility. Although the scientific understanding for the sign language is an essential step to build up a realistic recognition system, the proposed model can be used in other sign languages. The model exploits multi-stage Multiplicative Neural Networks for posture classification.

show abstract

Section: B Pcnn and Feature Generationmentioning

confidence: 99%

“…The McCulloch-Pitts model initiated the use of weighted summing units as the basic neuron model [20]. It is commonly known that there is only a minor closeness between this neuron model and the behavior of real biological neurons.…”

Section: Multi-stage Multiplicative Neural Network (Mmnn)mentioning

confidence: 99%

A Proposed Model for Standard Arabic Sign Language Recognition Based on Multiplicative Neural Network

Samir

Aboulela

Tolba

2014

The Egyptian Journal of Language Engineering

View full text Add to dashboard Cite

show abstract

“…For evaluating the performance of RFBP learning algorithm we proposed, the classifications of the non-convex in two dimensions (NC2) problem [16] are simulated. Three different learning methods are experimented for comparison, including conventional BP learning with constant parameters, stochastic BP learning and RFBP learning.…”

Section: Simulationsmentioning

confidence: 99%

“…In [16], a stochastic BP learning algorithm for dealing with the training problem of neural network is proposed by Chen et al A cost function is taken as the following form:…”

Section: Rfbp Learning Algorithmmentioning

confidence: 99%

An effective learning of neural network by using RFBP learning algorithm

Chen

Huang

Hwang

2004

Information Sciences

View full text Add to dashboard Cite

“…Drago et al have proposed an adaptive momentum BP for fast minimum search [5]. A randomized BP algorithm is proposed by Chen et al It is obtained by choosing a sequence of weighting vectors over the learning phase [6]. A new generalized BP algorithm is proposed in [7] to change the derivative of the activation function so as to magnify the backward propagated error signal, thus the convergence rate can be accelerated and the local minimum can be escaped.…”

Section: Introductionmentioning

confidence: 99%

Improved Back Propagation Algorithm to Avoid Local Minima in Multiplicative Neuron Model

Burse¹,

Manoria²,

Kirar³

2011

Information Technology and Mobile Communication

View full text Add to dashboard Cite

The back propagation algorithm calculates the weight changes of artificial neural networks, and a common approach is to use a training algorithm consisting of a learning rate and a momentum factor. The major drawbacks of above learning algorithm are the problems of local minima and slow convergence speeds. The addition of an extra term, called a proportional factor reduces the convergence of the back propagation algorithm. We have applied the three term back propagation to multiplicative neural network learning. The algorithm is tested on XOR and parity problem and compared with the standard back propagation training algorithm.

show abstract

A stochastic backpropagation algorithm for training neural networks

Cited by 7 publications

References 5 publications

A Proposed Model for Standard Arabic Sign Language Recognition Based on Multiplicative Neural Network

A Proposed Model for Standard Arabic Sign Language Recognition Based on Multiplicative Neural Network

An effective learning of neural network by using RFBP learning algorithm

Improved Back Propagation Algorithm to Avoid Local Minima in Multiplicative Neuron Model

Contact Info

Product

Resources

About