Skip to main content
U.S. flag

An official website of the United States government

NSCI Seminar: Mimicking Natural Ways of Computing in Solid-State Systems

While Boolean logic has been the backbone of information processing, there are computationally hard problems like solving constrained optimization problems wherein this conventional paradigm is fundamentally inadequate. This results in computational inefficacy, and motivates us to explore new pathways to their solution. In this talk, we introduce an experimental testbed comprising of compact coupled relaxation oscillator based dynamical system that exploits the insulator-metal transition in correlated oxide (for example VO2), to efficiently solve the vertex coloring of arbitrary graphs, a prototypical combinatorial optimization problem. Our work is inspired by the understanding that optimization finds a natural analogue in the energy minimization processes of highly parallel, coupled dynamical systems. Our work not only elucidates a physics-based computing method but also presents opportunities for building customized analog co-processors for solving computationally hard problems efficiently.

Sponsors

NSCI Committee

Created January 25, 2017