Testing Optimality of Sequential Decision-Making
M. Dörpinghaus, I. Neri, É. Roldán, H. Meyr and F. Jülicher
We introduce a statistical method to test whether a system that performs a binary sequential hypothesis test is optimal in the sense of minimizing the average decision times while taking decisions with given reliabilities. The proposed method requires samples of the decision times, the decision outcomes, and the true hypotheses, but does not require knowledge on the statistics of the observations or the properties of the decision-making system. We illustrate these tests with numerical simulations and discuss potential applications in machine learning and biology.
Records of entropy production in an electronic double dot
S. Singh, É. Roldán, I. Neri, I. M. Khaymovich, D. S. Golubev,
V. F. Maisi, J. T. Peltonen, F. Jülicher and J. P. Pekola
We measure single-electron fluctuations in a nanoelectronic device, a double dot, sealed at very low temperature 50mK.
Statistics of millions of records in the double dot confirm that entropy production’s average negative record cannot be below minus the Boltzmann constant. Moreover the data reveals a new bound for the maximal heat absorption of a nanoscopic system from its environment.
Thermodynamic Arcsine Law
A. C. Barato*, É. Roldán*, I. A. Martínez and S. Pigolotti
(* equal contribution)
We demonstrate with theory and experiment that the fraction of time a thermodynamic current elapses above its average value follows the arcsine law, a prominent result obtained by Levy for independent random variables.
Stochastic currents with long streaks above or below their average are much more likely than those that spend similar fraction of times above and below their average.
We demonstrate this result with simulations of molecular motors, quantum dots and colloidal systems, and with experimental data of a Brownian Carnot engine.
Generic properties of stochastic entropy production
S. Pigolotti, I. Neri, É. Roldán and F. Jülicher
Physical Review Letters 119 (14), 140604 (2017)
[ PDF ]
Introducing a random-time transformation, entropy production obeys a one-dimensional Ito drift-diffusion equation, independent of the underlying physical model.
This transformation leads to an exact uncertainty equality between the Fano factor of entropy production and the Fano factor of the entropic time.
Colloidal heat engines: A review
[ PDF ]
We review recent experimental constructions of small heat engines which are the colloidal equivalents to the macroscopic Stirling, Carnot and steam engines, whose fluctuations lead to unique phenomena that have no equivalent in the macroscopic world. We also discuss the work extraction from bacterial reservoirs.
Finally, we provide some guidance on how the work extracted from colloidal heat engines can be used to generate net particle or energy currents, proposing a new generation of experiments with colloidal systems.
Path-integral formalism for stochastic resetting:
Exactly solved examples and shortcuts to confinement
É. Roldán and S. Gupta
Physical Review E 96 (2), 022130 (2017)
[ PDF ]
We present a path-integral approach to derive analytical expressions for a variety of statistics of the dynamics of overdamped Brownian particles under stochastic resetting.
We find that using energy-dependent resetting processes is more effective in achieving spatial confinement of Brownian particles on a faster time scale than performing quenches of parameters of the harmonic potential.