Information Dynamics at the Edge of Chaos

Many evolutionary and self-organization pressures can be characterized information theoretically not only because it's an approximation useful in designing biologically-inspired systems, but also because numerous optimal structures evolve/self-organize in nature when information dynamics approach critical points. The talk will focus on information dynamics of computation within spatio-temporal systems in terms of three fundamental operations: information storage, transfer, and modification, quantifying these operations on a local scale in space and time. The methods will be exemplified in different contexts, including swarms, computational neuroscience, and random Boolean networks. In addition, we shall discuss a relation between Fisher information and phase transitions / order parameters, drawing from both thermodynamics and statistical estimation theory, as well as a thermodynamic interpretation of transfer entropy.