Papers
Topics
Authors
Recent
Search
2000 character limit reached

Concentration of additive functionals of Stratonovich-type

Published 1 Sep 2026 in math.PR, cond-mat.stat-mech, and math-ph | (2609.01581v1)

Abstract: Additive functionals J<em>t=1t0<sup>tU(Xs)</sup>dXs\overline{J}<em>t=\frac{1}{t}\int_0<sup>tU(X_s)\circ</sup> dX_s of Stratonovich-type recently attracted much attention in the context of inference of thermodynamic properties of complex systems from observations UU of individual fluctuating paths (Xs)</em>0st(X_s)</em>{0\le s\le t}, whereby X0X_0 is initiated from some general measure. Concentration results on Jt\overline{J}_t, albeit desirable, are virtually nonexistent. They turn out to be significantly more challenging to prove than for classical Lebesgue-type functionals ρt=1t0<sup>t</sup>V(Xs)ds\overlineρ_t=\frac{1}{t}\int_0<sup>t</sup> V(X_s)ds because the tilt deforms the full second-order structure of the Feynman-Kac generator instead of contributing an additive potential. This renders the generator generally non-self-adjoint even under detailed balance. We overcome this by working with a symmetrized Dirichlet form with a new effective potential that now couples the observable to the non-equilibrium character of the dynamics. We prove concentration inequalities for Jt\overline{J}_t for any bounded, sufficiently smooth vector-valued function UU of a general geometrically ergodic diffusion process XsX_s, including explicit sub-gamma and Bernstein-type inequalities, and we obtain explicit upper bounds on Var(Jt){\rm Var}(\overline{J}_t). Strikingly, under detailed balance the concentration of Jt\overline{J}_t is distinctively sub-Gaussian at all times and all deviations, with a variance proxy fixed by the noise alone and independent of the spectral gap, which has no analog for ρt\overlineρ_t.

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.