Theoretical investigations of an information geometric approach to complexity

摘要: It is known that statistical model selection as well identification of dynamical equations from available data are both very challenging tasks. Physical systems behave according to their underlying which, in turn, can be identified experimental data. Explaining requires selecting mathematical models best capture the regularities. The existence fundamental links among physical systems, equations, and modeling motivate us present this article our theoretical scheme which combines information geometry inductive inference methods provide a probabilistic description complex presence limited information. Special focus devoted describe role entropic geometric complexity measure. In particular, we several illustrative examples wherein used infer macroscopic predictions when only partial knowledge microscopic nature given system available. Finally, limitations, possible improvements, future investigations discussed.