Neural lab a simulator for artificial neural networks
Sergio Ledesma, Mario-Alberto Ibarra-Manzano, M. G. García-Hernández, Dora-Luz Almanza-Ojeda · 2017 Computing Conference · 2017
Artificial neural networks are inspired by biologic processes. Artificial neural networks are important because they can be used to deduct a function from observations, in other words artificial neural networks can learn from experience. This paper introduces a simulator called Neural Lab for artificial neural networks. Neural Lab is implemented using object-oriented programming by a set of C++ classes and these classes can be used to create a standalone application with artificial neural networks. Several optimization techniques were applied in the design and implementation of this simulator. This simulator provides support for multi-layer feed forward networks, probabilistic neural networks and Kohonen networks. Finally, in order to test the proposed simulator, several computer experiments were performed. These tests included mapping problems and auto-associate problems.