2021 BRN Discussion, page-33309

On the theme from yesterday as to what cannot AKIDA technology not do? I have no idea but here are two things it can do:

I. INTRODUCTION The main question that the paper addresses is: Can the third generation of neural networks – spiking neural networks (SNN) be used to model and understand dynamic, spatio-temporal, cognitive processes in the brain? A question to follow would be: Can this approach be translated into intelligent robotic systems? The paper argues that, SNN can be used to model data that represent brain, spatio-temporal cognitive processes. Such models can be further implemented as neuromorphic cognitive systems using the latest neuromorphic hardware platforms.

The paper proposes new algorithms for encoding, learning and classification of functional magnetic resonance imaging (fMRI) data that measure dynamic cognitive processes. The algorithms are part of the recently proposed NeuCube SNN architecture. The model is illustrated on two case study fMRI data related to seeing a picture versus reading a sentence.

https://scholar.google.com.au/scholar_url?url=https://pure.ulster.ac.uk/ws/files/79064909/IEEE_Transactions_on_Cognitive_and_Developmental_Systems.pdf&hl=en&sa=X&ei=CPPLYZbhEaKbywSgloz4Ag&scisig=AAGBfm0W6X2g39Dx9eqEVVIQRlO2rncBew&oi=scholarr

Abstract—Synergies between wireless communications and artificial intelligence are increasingly motivating research at the intersection of the two fields. On the one hand, the presence of more and more wirelessly connected devices, each with its own data, is driving efforts to export advances in machine learning (ML) from high performance computing facilities, where information is stored and processed in a single location, to distributed, privacy-minded, processing at the end user. On the other hand, ML can address algorithm and model deficits in the optimization of communication protocols. However, implementing ML models for learning and inference on batterypowered devices that are connected via bandwidth-constrained channels remains challenging. This paper explores two ways in which Spiking Neural Networks (SNNs) can help address these open problems. First, we discuss federated learning for the distributed training of SNNs, and then describe the integration of neuromorphic sensing, SNNs, and impulse radio technologies for low-power remote inference.

https://arxiv.org/pdf/2010.14220

These are not my opinions but the opinions of independent researchers exploring the limits of SNN so DYOR
FF

AKIDA BALLISTA