Spatial computing in structured spiking neural networks with a robotic embodiment

Lobov, Sergey; Mikhaylov, A. N.; S., Berdnikova, Ekaterina; Makarov, Valeri A.; Kazantsev, Victor B.

doi:10.48550/arxiv.2112.07150

Cited by 1 publication

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the implementations of this concept can be considered a learning model in a neural network based on the STDP association of interacting traveling waves ( Alexander et al, 2011 ; Palmer and Gong, 2014 ). Spatial computing in small neural circuits and modular SNNs can simulate Pavlovian conditioning and operant learning in neurorobots ( Lobov et al, 2020b , 2021a ). Another possible way to implement spatial computations is cognitive maps and spatial memory with positive ( Ponulak and Hopfield, 2013 ) or negative ( Lobov et al, 2021b ) environmental stimuli ( Figure 1C ).…”

Section: Spiking Neural Network As An Alternative For Building Reflec...mentioning

confidence: 99%

“…Another possible way to implement spatial computations is cognitive maps and spatial memory with positive ( Ponulak and Hopfield, 2013 ) or negative ( Lobov et al, 2021b ) environmental stimuli ( Figure 1C ). Note that due to the presence of spontaneous activity in SNNs (unlike ANNs), they can “live” without external input, determining the “behavior” of neurorobots ( Lobov et al, 2020b , 2021a , 2021b ).…”

Section: Spiking Neural Network As An Alternative For Building Reflec...mentioning

confidence: 99%

See 1 more Smart Citation

Toward Reflective Spiking Neural Networks Exploiting Memristive Devices

Makarov

Lobov

Shchanikov

et al. 2022

Front. Comput. Neurosci.

Self Cite

View full text Add to dashboard Cite

The design of modern convolutional artificial neural networks (ANNs) composed of formal neurons copies the architecture of the visual cortex. Signals proceed through a hierarchy, where receptive fields become increasingly more complex and coding sparse. Nowadays, ANNs outperform humans in controlled pattern recognition tasks yet remain far behind in cognition. In part, it happens due to limited knowledge about the higher echelons of the brain hierarchy, where neurons actively generate predictions about what will happen next, i.e., the information processing jumps from reflex to reflection. In this study, we forecast that spiking neural networks (SNNs) can achieve the next qualitative leap. Reflective SNNs may take advantage of their intrinsic dynamics and mimic complex, not reflex-based, brain actions. They also enable a significant reduction in energy consumption. However, the training of SNNs is a challenging problem, strongly limiting their deployment. We then briefly overview new insights provided by the concept of a high-dimensional brain, which has been put forward to explain the potential power of single neurons in higher brain stations and deep SNN layers. Finally, we discuss the prospect of implementing neural networks in memristive systems. Such systems can densely pack on a chip 2D or 3D arrays of plastic synaptic contacts directly processing analog information. Thus, memristive devices are a good candidate for implementing in-memory and in-sensor computing. Then, memristive SNNs can diverge from the development of ANNs and build their niche, cognitive, or reflective computations.

show abstract