English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Artificial neural networks (ANNs) are computational models of their biological counterparts. They consists of densely interconnected computing units that work together to solve a specific problem. The information, which is acquired during a learning process, is stored in the synaptic weights of the internodal connections. The main advantage of neural networks is their ability to represent complex functions and the efficient storage of information. ANNs are frequently employed in applications involving data classification, function approximation, and signal processing (Haykin, 1994). The topology of ANNs consists of an arrangement of neurons, which are equipped with a transfer function and synaptic weights, and the nodal connections. Despite these simple topological elements, the flexible arrangement of neurons and connections allows the generation of ANNs with arbitrary complexity. The resulting topological complexity, however, directly affects the network performance. The performance, or fitness, is a measure of the accuracy of a network in representing an input-output relation. For instance, network topologies with only few neurons and synaptic weights provide only limited flexibility in representing complex functions. They have therefore typically only a poor fitness. On the other side, complex networks that provide large flexibility in representing new data, can lead to poor generalizability and extensive computational costs for training and data retrieval (Yao, 1999). Considering the integration of such networks in large-scale simulations, data retrieval from such large networks can lead to a significant increase in overall computing time. Because of the inherent topological complexity it is apparent that the a priori identification of a network topology with near-optimal performance is a challenging task, and is often guided by heuristics or trial-and-error. The design of a specific network topology with optimal performance can be formulated as an optimization problem. The choice of the method to solve this problem is determined by the inherent properties of the ANN (Miller et al., 1989):

intech ebooks

Construction of Optimal Artificial Neural Network Architectures for Application to Chemical Systems: Comparison of Generalized Pattern Search Method and Evolutionary Algorithm