Controlling stochastic logic circuits subject to thermodynamic tradeoffs

Landauer’s principle and developments from stochastic thermodynamics such as the thermodynamic uncertainty relations provide bounds on the energy costs of logical operations. As transistors on chips become smaller, one must begin to consider the effects of thermal fluctuations that approach the magnitude of the correspondingly smaller operating voltages. We explore tradeoffs among speed, accuracy, and dissipation in a thermodynamically consistent model of complementary logic gates to understand the fundamental bounds and design principles for computing near thermal energy scales. From the basic building block of a NOT gate, we can construct examples of larger systems such as a memory storage device and clock, yielding insight into how thermodynamic tradeoffs might scale to more complex architectures. With an understanding of thermodynamic tradeoffs, we explore how to control these noisy systems in order to maximize operating certainty and efficiency in these systems.