Level 2.5 large deviations and uncertainty relations for non-Markov self-interacting dynamics

by Dr Francesco Coghi (Univ. Nottingham)

Thursday 2 Apr 2026, 11:30 → 12:30 Europe/Rome

1/1-2 - Aula "C. Voci" (Dipartimento di Fisica e Astronomia - Edificio Marzolo)

Self-interacting jump processes are stochastic systems where transition rates depend on the process' own empirical occupation measure—the  time-averaged distribution over visited states—introducing a feedback that breaks Markovianity. In this talk, I will present a large deviation framework for such systems. I will derive the level-2.5 rate functional for the joint fluctuations of occupation measure and probability flux, constructed via exponential tilting extended to the non-Markovian setting. I will then show how memory modifies fluctuation bounds, yielding kinetic and thermodynamic uncertainty relations that generalise their Markov counterparts. Depending on time, I will close with one or two examples illustrating how feedback reshapes fluctuation-dissipation trade-offs.