PGI-1 Talk: Michael Dick
Renormalized Fluctuation Expansion for Non-Equilibrium Disordered Networks
- begin
- 21 Oct 2020 11:30
- venue
- BigBlueButton
Abstract:
Critical as well as chaotic behavior in cortical networks have both been suggested to play an important role but are often used interchangeably.
A point close to criticality leads to what is known as critical slowing down and thus large time constants of correlations which offer a mechanism for memory.
A point close to the transition to/ at the onset of chaos is said to be optimal for storage, transmission and modification of information.
To study the interplay of both behaviors we consider an scs-model with a nonzero mean connectivity but as the classical Mean-Field calculations are not able to describe the second order phase transitions taking place or behavior close to criticality, we use Legendre transformation methods which circumvents these shortcomings.
We derive a set of self-consistent equations for the first two renormalized green's functions and a way to solve them numerically as well as provide analytical solutions for a linear network.