7 papers · 1 filter
Speed Limit for Information Acquisition in Stochastic Learning Dynamics
Shuta Kobayashi, Andreas Dechant
Neural networks acquire internal representations through learning. In this work, we formulate stochastic gradient descent (SGD) as a Markovian stochastic process and derive a Fishe…
Renormalized entropy production for optimal transport in jump processes: Make conservative forces optimal again
Andreas Dechant, Jann van der Meer
For continuous-space diffusion processes, there is a strong connection between conservative forces and entropy production. For a given time evolution of the system's state, the ent…
Markovian thermodynamics of non-Markovian Langevin equations
Andreas Dechant, Kiyoshi Kanazawa
We develop the thermodynamics of non-Markovian generalized Langevin equations by embedding them in a high-dimensional Markovian representation involving auxiliary degrees of freedo…
Near-optimality of conservative driving in discrete systems
Jann van der Meer, Andreas Dechant
Transferring a physical system from an initial to a final state while minimizing energetic losses is an interdisciplinary control problem that bridges stochastic thermodynamics and…
Entropy production and non-Gaussianity of fast processes at weak damping
Mario A. Ciampini, Jakob Rieser, Nikolai Kiesel +1
We present a method of estimating the rate of entropy production in underdamped dynamics by decomposing it into contributions originating in different non-equilibrium effects. Spec…
Finite-frequency fluctuation-response inequality
Andreas Dechant
We derive an inequality relating the finite-frequency linear response and fluctuations of an observable in a physical system. The relation holds for arbitrary observables and pertu…