TOWARDS A THEORETICAL BASIS FOR MODELLING OF HIDDEN LAYER ARCHITECTURE IN ARTIFICIAL NEURAL NETWORKS
Published In: 2ND INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, ELECTRONICS AND COMMUNICATION
Author(s): ASOKA S. KARUNANANDA , N. MIHIRINI WAGARACHCHI
Abstract: Artificial neural networks (ANNs) are mathematical and computational models that are inspired by the biological neural systems. Just like biological neural networks become experts by learning from the surrounding, ANNs also have the ability to be experts in the particular area by training the network. Despite of their many advantages, there are some unsolved problems in applying artificial neural networks. Determine the most efficient architecture for the given task is identified as one of those major issues. This paper provides a pruning algorithm based on the backpropagation training algorithm to obtain the optimal solution of ANN. The pruning is done according to the synaptic pruning in biological neural system. Experiments were done with some well known problems in machine learning and artificial neural networks and results show that the new model performs better than the initial network in training data sets.
- Publication Date: 26-Oct-2014
- DOI: 10.15224/978-1-63248-029-3-73
- Views: 0
- Downloads: 0
AN ANALYSIS OF SUBFIELD BASED ON FUZZY FOR DETERMINISTIC INDOOR LOCALIZATION IN WSN
Published In: 2ND INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, ELECTRONICS AND COMMUNICATION
Author(s): GUNGOR YILDIRIM , YETKIN TATAR
Abstract: In many wireless applications, location awareness is of paramount importance and, therefore, there are many algorithms regarding localization problems in literature. Fingerprint-based algorithms are one of them, consisting of two phases; mapping and location estimation. In location estimation phases of some technics based on fingerprint, subfields are used for the purpose of filtering data. A subfield analysis could be complex in applications having big radio maps. In addition to this, unstable RSSI values could make the analysis more difficult. In such a state, different approaches can be used to increase the efficiency. In this paper, an alternative deterministic localization technique suggested by the authors of this paper in an earlier study was explained and then unlike the earlier study, a subfield analysis based on fuzzy was applied with the aim of examining a soft computing approach in localization. The results of the experiments were compared with the results of classical dete
- Publication Date: 26-Oct-2014
- DOI: 10.15224/978-1-63248-029-3-74
- Views: 0
- Downloads: 0