The cascade-correlation learning architecture
Advances in neural information processing systems 2
Neural networks: algorithms, applications, and programming techniques
Neural networks: algorithms, applications, and programming techniques
Constraint satisfaction problems solved by semidefinite relaxations
WSEAS Transactions on Computers
Neural networks in production control
CSCC'11 Proceedings of the 2nd international conference on Circuits, Systems, Communications & Computers
| Hi-index | 0.00 |
Current platforms for the definition and operation of artificial neural networks are oriented towards a specialized user profile. This article proposes a series of software components that provide abstraction in platform development, both with regard to the implementation of the components and the operation of the network that was chosen to resolve a problem. This proposal has the ambitious purpose of minimizing the learning curve of the tools and providing a framework that is oriented towards a more generic user profile. The framework is composed by three systems: a library, in which we factorize the functionalities related to the networks, and two tools, an interpreter and aWeb application, which provide two different points of view for the development of the networks, but both rely on the library to provide the features related to the ANNs.