In many real world processes, the evolution of the system depends not only on its current state, but also on its prehistory. The hysteresis law can describe a wide class of such dependencies arising in physical, biological, economic, and financial applications. In this project, we develop specific architectures of artificial neural networks that take into account underlying hysteretic processes and allow one to effectively model the evolution of the system.

Principle Investigators: Pavel Gurevich, Hannes Stuke
Members: Nikita Begun, Zongxiong Chen

Collaboration partner:
Prof. Dimitri Rachinskii (University of Texas at Dallas))