Parameter synthesis of polynomial dynamical systems

Parametric dynamical systems emerge as a formalism for modeling natural and engineered ranging from biology, epidemiology, medicine to cyber-physics. Parameter tuning is complex task usually performed exploiting heavy simulations having high computational complexity not ensuring the correctness of synthesized systems. We consider problem parameter synthesis discrete-time polynomial propose formal method based on Bernstein coefficients that allows refining set parameters according temporal specification defined Signal Temporal Logic formula. The system correct with respect we demonstrate scalability approach by implementing it in C++ library Sapo, also available within stand-alone application web application. Finally, illustrate tool usage interface through simple yet realistic epidemiological model an intriguing enhancing accuracy verification neural networks.