Page 141 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

130 CHAPTER 6 Evolving and Spiking Connectionist Systems

FIGURE 6.12
Deep learning of seismic data: (A) A map of the seismic stations in New Zealand; (B)
Connectivity of part of a trained SNNcube on the seismic data from New Zealand stations
[118,133].

5. CONCLUSION
Artificial Neural Networks (ANN) and SNN are the latest development in the history
of AI, and perhaps the most contemporary and promising techniques nowadays
towards building brain-inspired AI. The current development of ANN and more
specifically ECOS and SNN is linked tightly with other areas of computational
intelligence, such as logic and rule based systems, evolutionary computation, and
very much with neuroscience. The future progress of ANN and SNN will be part
of the progress in the broad areas of science in general and understanding the other
related areas is essential. The chapter presents principles of AI methods and ANN
and then surveys their use in practical systems. The chapter specifically addresses
the problem of designing evolving AI systems for adaptive learning and knowledge
discovery from complex data, emphasizing on evolving connectionist systems
(ECOS). Different types of ECOS methods are suitable for different applications
as discussed in the chapter. Spiking neural networks (SNN) are discussed as
promising brain-inspired techniques for brain-like AI applications. One of
the SNN platforms, NeuCube is presented for deep learning and deep knowledge
representation of temporal or spatiotemporal data with some case study applications
on fMRI and seismic data. The chapter advocates that knowing and integrating