Demystifying Nonequilibrium Statistical Mechanics
Recent general results on the statistics of nonequilibrium processes have opened up old debates between the exact dynamical and informational viewpoints on probability. Many of the good properties of equilibrium systems are not rigorously provable without assuming ergodicity. It turns out those arguments are even more relevant, and more pernicious, when working in a dynamical context. Even though nonequilibrium research predates traditional equilibrium thermodynamics, it is still seen by many as a vast, uncharted territory. In this talk, I show how there is a growing consensus among different camps about the general form for framing problems in nonequilibrium statistical mechanics. This provides three independent derivations of the same general formula. Applications to a few diverse classical nonequilibrium problems argue that we can even carry over most of the maximum entropy ideas underlying the equilibrium theory. Finally, I show how classical statistical mechanics emerges from a quantum nonequilibrium model without any weak-coupling assumptions.