Work

Natural Dynamics for Combinatorial Optimization

Year: 2015
Type: preprint
Abstract: Stochastic and or natural dynamical systems (DSs) are dominated by sudden nonlinear processes such as neuroavalanches, gamma-ray bursts, solar flares, earthquakes etc. that exhibit scale-free statistics. These behaviors also occur in many nanosystems. On phase diagrams, these DSs belong to a finite-width phase that separates the phases of thermodynamic equilibrium and ordinary chaotic dynamics, and that is known under such names as intermittency, noise-induced chaos, and self-organized criticality. Within the recently formulated approximation-free cohomological theory of stochastic differential equations, the noise-induced chaos can be roughly interpreted as a noise-induced overlap between regular (integrable) and chaotic (non-integrable) deterministic dynamics so that DSs in this phase inherit the properties of the both. Here, we analyze this unique set of properties and conclude that such DSs must be the most efficient natural optimizers. Based on this understanding, we propose the method of the natural dynamical optimization using DSs at the phase of the noise-induced chaos. This method can be viewed as a hybrid of the simulated and chaotic annealing methods. Our proposition can result in a new generation of (hardware) devices for the efficient solution of various search and or combinatorial optimization problems.
Authors: I. V. Ovchinnikov, Kang L. Wang
Language: English
Open Access status: green
OpenAlex ID: W2252336880
Cites: 24
Cited by:
Related to: 10