A New Back-Propagation Neural Network Optimized with Cuckoo Search Algorithm

Nawi, Nazri Mohd; Khan, Abdullah; Rehman, Muhammad Zubair

doi:10.1007/978-3-642-39637-3_33

Cited by 66 publications

(33 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A well-trained ANN should consist of the optimal numbers of hidden layer neurons, hidden layer(s), and weight values, which sufficiently avoid the risk of either under-or overfitting. In practical applications, there is a large number of neural networks with modified algorithms, such as ELM [43][44][45], backpropagation neural network (BPNN) [46][47][48], and general regression neural network (GRNN) [49][50][51]. Though there are various network models, the basic principles for model training are similar.…”

Section: Machine Learning Methodsmentioning

confidence: 99%

Predictive Power of Machine Learning for Optimizing Solar Water Heater Performance: The Potential Application of High-Throughput Screening

Liu

et al. 2017

International Journal of Photoenergy

View full text Add to dashboard Cite

Predicting the performance of solar water heater (SWH) is challenging due to the complexity of the system. Fortunately, knowledge-based machine learning can provide a fast and precise prediction method for SWH performance. With the predictive power of machine learning models, we can further solve a more challenging question: how to cost-effectively design a highperformance SWH? Here, we summarize our recent studies and propose a general framework of SWH design using a machine learning-based high-throughput screening (HTS) method. Design of water-in-glass evacuated tube solar water heater (WGET-SWH) is selected as a case study to show the potential application of machine learning-based HTS to the design and optimization of solar energy systems.

show abstract

Section: Machine Learning Methodsmentioning

confidence: 99%

Predictive Power of Machine Learning for Optimizing Solar Water Heater Performance: The Potential Application of High-Throughput Screening

Liu

et al. 2017

International Journal of Photoenergy

View full text Add to dashboard Cite

show abstract

“…For each time step t, the movement of the virtual bats is given by updating their velocity and position using Eqs. (5)- (7). From the previous studies on Bat algorithm, it was realized that whenever the Bat deals with lower-dimensional optimization problems, it gives good results.…”

Section: The Proposed Bagd Algorithmmentioning

confidence: 97%

A Modified Bat Algorithm Based on Gaussian Distribution for Solving Optimization Problem

Nawi

Rehman

Khan

et al. 2016

j comput theor nanosci

Self Cite

View full text Add to dashboard Cite

An improved Bat algorithm with Gaussian distribution random walk (BAGD) is introduced in this paper. The original Bat algorithm has a problem of random large step length that leads to suboptimal solutions in the search space and it cannot solve higher dimensional problems. To solve higher dimensional problems and to decrease the step length size, this research focuses on using a Gaussian distribution in Bat algorithm which provide shorter step lengths during the search. The proposed BAGD was compared with six popular metaheuristic algorithms on ten benchmark functions. Comparative results indicated that the proposed BAGD perform better than the stateof-the-art algorithms in most cases. The proposed BAGD solution used small step lengths in the search space and it was able to solve high dimensional problems.

show abstract

“…Some hybrid variants of back-propagation have been proposed in order to take advantage of the benefits from other algorithms. For example, combining it with cuckoo search algorithm can increase the searching speed for optimal solutions [96]. The combination with ant colony algorithm can decrease the computational cost with increased stability of convergence [97].…”

Section: Gradient Decent Optimizationmentioning

confidence: 99%

Recent advances in convolutional neural network acceleration

et al. 2019

View full text Add to dashboard Cite

In recent years, convolutional neural networks (CNNs) have shown great performance in various fields such as image classification, pattern recognition, and multi-media compression. Two of the feature properties, local connectivity and weight sharing, can reduce the number of parameters and increase processing speed during training and inference. However, as the dimension of data becomes higher and the CNN architecture becomes more complicated, the endto-end approach or the combined manner of CNN is computationally intensive, which becomes limitation to CNN's further implementation. Therefore, it is necessary and urgent to implement CNN in a faster way. In this paper, we first summarize the acceleration methods that contribute to but not limited to CNN by reviewing a broad variety of research papers. We propose a taxonomy in terms of three levels, i.e. structure level, algorithm level, and implementation level, for acceleration methods. We also analyze the acceleration methods in terms of CNN architecture compression, algorithm optimization, and hardware-based improvement. At last, we give a discussion on different perspectives of these acceleration and optimization methods within each level. The discussion shows that the methods in each level still have large exploration space. By incorporating such a wide range of disciplines, we expect to provide a comprehensive reference for researchers who are interested in CNN acceleration.

show abstract

A New Back-Propagation Neural Network Optimized with Cuckoo Search Algorithm

Cited by 66 publications

References 30 publications

Predictive Power of Machine Learning for Optimizing Solar Water Heater Performance: The Potential Application of High-Throughput Screening

Predictive Power of Machine Learning for Optimizing Solar Water Heater Performance: The Potential Application of High-Throughput Screening

A Modified Bat Algorithm Based on Gaussian Distribution for Solving Optimization Problem

Recent advances in convolutional neural network acceleration

Contact Info

Product

Resources

About