Udo Seifert

TL;DR: Using stochastic thermodynamics, it is shown that the thermodynamic costs of the learning process provide an upper bound on the amount of information that the network is able to learn from its teacher for both batch and online learning.

TL;DR: This manuscript considers a single protein (elastic spring of a finite rest length) pinning a membrane modeled in the Monge gauge and explores static correlations of the free and the pinned membrane, as well as the membrane shape, showing that all three are mutually interdependent and have an identical long-range behavior characterized by the correlation length.

TL;DR: It is shown numerically that even in a moderately dense suspension the latter is negligible, and the fluctuation-dissipation theorem is approximately derived in a "hybrid" form involving the kinetic temperature as an effective temperature and an additive correction term.

TL;DR: An effective Monte Carlo scheme within which the effects of the membrane are integrated into local rates for molecular recognition is presented, which yields detailed information about protein transport and complexation in membranes, a fundamental step toward understanding even more complex membrane interactions in the cellular context.

TL;DR: It is shown that the optimized model leads to more work extraction in comparison to the memory-less model, with the gain parameter being larger in the region where the frequency of non-reliable measurements is higher and the model has two second law inequalities.