Exploiting a Stimuli Encoding Scheme of Spiking Neural Networks for Stream Learning
This work addresses stream data processing challenges for applications in big data scenarios, but it is incremental as it builds on existing encoding schemes.
The study tackled the problem of improving stream learning performance by focusing on the Gaussian receptive fields in the population encoding scheme of Spiking Neural Networks, applying them as a pre-processing technique to datasets, and experiments confirmed successful application in many real cases with boosted predictive performance.
Stream data processing has gained progressive momentum with the arriving of new stream applications and big data scenarios. One of the most promising techniques in stream learning is the Spiking Neural Network, and some of them use an interesting population encoding scheme to transform the incoming stimuli into spikes. This study sheds lights on the key issue of this encoding scheme, the Gaussian receptive fields, and focuses on applying them as a pre-processing technique to any dataset in order to gain representativeness, and to boost the predictive performance of the stream learning methods. Experiments with synthetic and real data sets are presented, and lead to confirm that our approach can be applied successfully as a general pre-processing technique in many real cases.