|author(s)||A. Baumbach, A. F. Kungl, M. A. Petrovici, J. Schemmel, K. Meier|
|title||Magnetic Phenomena in Spiking Neural Networks|
|Keywords||Statistical Physics, Non-linear systems, Neural Networks|
|source||Fruehjahrstagung der DPG|
Systems close to criticality are always of particular interest. The arguably simplest model known to exhibit critical phenomena is the Ising model for ferromagnetism. Recent work on spiking neural networks developed a description of these biologically inspired networks under poissonian noise input as a Boltzmann machine. As such a description is widely used in neuroscience to effectively describe biological models and data one would expect that all the phenomena known from statistical physics can also be observed in these systems.
This work investigates a simplified model, the Neuralsampling framework introduced by Buesing et al., which we modify to include exponentially decaying interactions (resembling biological interactions) in an Ising-like network. While the global properties, like the unordered phase in the infinite temperature limit and the ordered phase in the zero temperature limit, we show that the phase diagram of this model shows richer phenomena than the classical Ising model. For example it allows a system to pass through multiple phases, rather than only the unordered-ordered transition, while cooling down.