Elsevier

Neurocomputing

Volume 164, 21 September 2015, Pages 1-4
Neurocomputing

Editorial
Advances in computational intelligence: Selected and improved papers of the 12th International Work-Conference on Artificial Neural Networks (IWANN 2013)

https://doi.org/10.1016/j.neucom.2015.02.074Get rights and content

Section snippets

Machine learning methods and techniques

The paper Maximum Margin Clustering for State Decomposition of Metastable Systems, by Hao Wu, proposes a method to cluster the phase space of a dynamical system into a set of metastable regions. In the permanent regime, the system will remain at one of such metastable states, before a transition to a different region occurs, so that two very different time scales arise from considering movements within and between metastable states. The notion of maximum margin clustering is adapted to provide

Continuous-time methods

Yadira Hernández-Solano et al. focus on the construction of numerical methods for implementing gradient algorithms, as presented in their paper entitled A discrete gradient method to enhance the numerical behaviour of Hopfield networks. Machine learning algorithms defined in continuous time, such as Hopfield neural networks, are often modelled by systems of ordinary differential equations, which must be numerically approximated to allow their computer implementation. This paper presents a

Data mining and image processing

Eduardo de la Hoz et al. tackle the problem of computer intrusion detection in their paper PCA filtering and Probabilistic SOM for Network Intrusion Detection. A hybrid system is designed to distinguish normal network traffic from anomalous connections that represent a risk to the system or are unauthorized. First of all, connection data are filtered by means of Principal Component Analysis, together with a feature selection method based upon the Fisher discriminant ratio. Then, a fuzzy version

Biological applications

In the paper Comparing different machine learning and mathematical regression models to evaluate multiple sequence alignments, Francisco M. Ortuño et al. propose a methodology to evaluate multiple sequence alignments that result from comparing molecular chains. Four regression algorithms, namely Regression Trees, Bootstrap Aggregation Trees, Least-Squares Support Vector Machines, and Gaussian Processes, are assessed, providing a significant improvement in terms of correlation with respect to

References (24)

  • A. Prieto et al.

    Computational intelligence and bioinspired systems

    Neurocomputing

    (2007)
  • J. Cabestany et al.

    Computational and ambient intelligence

    Neurocomputing

    (2009)
  • F. Sandoval et al.

    Bio-inspired systems: computational and ambient intelligence

    Neurocomputing

    (2011)
  • A. Prieto et al.

    Advances in artificial neural networks and machine learning

    Neurocomputing

    (2013)
  • A. Prieto (Ed.), Artificial neural networks, International Workshop, IWANN ׳91, Granada, Spain, September 17–19, 1991,...
  • J. Mira, J. Cabestany, A. Prieto (Eds.), New Trends in Neural Computation: International Workshop on Artificial Neural...
  • J. Mira, F. Sandoval (Eds.), From Natural to Artificial Neural Computation: International Workshop on Artificial Neural...
  • J. Mira, R. Moreno-Díaz, J. Cabestany (Eds.), Biological and Artificial Computation: from neuroscience to technology:...
  • J. Mira, J. Sánchez-Andrés (Eds.), Foundations and Tools for Neural Modeling. International Work-Conference on...
  • J. Mira, J. Sánchez-Andrés (Eds.), Engineering Applications of Bio-Inspired Artificial Neural Networks. International...
  • J. Mira, A. Prieto (Eds.), Connectionist Models of Neurons, Learning Processes, and Artificial Intelligence, Sixth...
  • J. Mira, A. Prieto (Eds.), Bio-Inspired Applications of Connectionism, Sixth International Work-Conference on...
  • View full text