In conventional computing, the basic unit of information, i.e., a “bit”, is stored as a “0” or “1” in deterministic units such as MOS transistors or stable memory elements. The probabilistic computing paradigm is based on a “p-bit” that randomly fluctuates between 0 and 1, a behavior that is mimicked by the physics of low barrier nanomagnets. We present how a wide variety of computing problems, traditionally solved in software, can be solved in hardware by coupling such nanomagnets through spin currents to form correlated networks. Such a network of coupled stochastic nanomagnets traverses through its collective states and is naturally guided by the laws of statistical mechanics towards the low energy states that represent the optimal solution. We will discuss the current experimental realizations of p-bits and p-circuits and outline the path ahead.
Punyashloka Debashis obtained his B.Tech from IIT Kanpur in 2012 and M.Tech from IIT Bombay in 2014, both in Electrical Engineering. He joined Purdue University in Fall 2014 and is currently pursuing his PhD degree in the School of Electrical and Computer Engineering. His research interests include exploring spin as the new state variable for logic computing.