Exact Fluctuation Theory for Product-Kernel Aggregation, Instantons, Non-Convex Large.. by R. Rajesh

Exact Fluctuation Theory for Product-Kernel Aggregation, Instantons, Non-Convex Large.. by R. Rajesh

Title history
  • Exact Fluctuation Theory for Product-Kernel Aggregation, Instantons, Non-Convex Large.. by R. Rajesh — observed on 20261106-0125 UTC (YouTube)
Formal & Physical Sciences Physics PHPhysicsPHUMathematical
🎙 R. Rajesh 👥 75K 📅 November 5, 2026 ⏱ 30 min 👁 340 📄 original study 🧭 2026-11-06
Available in: English (current) Français

Keywords

product kernelgelationcoagulationcanonical ensemblephase diagram

Summary

The talk, presented by R. Rajesh at the ICTS program on Macroscopic Fluctuation Theory, addresses exact results for the probability distribution of the number of particles in a product-kernel aggregation model. The speaker derives the complete large deviation function (LDF) including non-convex parts, using both canonical and grand-canonical approaches. They identify a regime where the cumulant generating function becomes non-differentiable, leading to ensemble inequivalence. The optimal trajectories (instantons) are computed analytically via the Doi-Peliti-Zeldovich method, revealing a dynamical gelation transition along the rare-event paths. The phase diagram in the plane of scaled time and final density exhibits a continuous transition and a first-order lines with a coexistence region, along with spinodal lines. Numerical simulations match the theoretical predictions for monomer density, dimer density, and full mass distribution. The work extends to generic kernels and open questions are discussed.

138 words

Critical Evaluation

Value of the Information & Strength of the Argument

The presentation offers a substantial theoretical contribution by providing exact, non-perturbative results for a classic stochastic process. The argumentation is rigorous: the speaker derives explicit formulas for the generating function and large deviation function, demonstrates non-convexity via exact numerical enumeration, and validates the instanton calculations with Monte Carlo simulations. The clarity of the logical progression—from model definition to exact solution to phase diagram—is compelling. The claim of ensemble inequivalence is substantiated by explicitly showing that the grand-canonical Legendre transform yields only the convex envelope, obscuring the full LDF. The discussion of spinodal lines and metastability is grounded in the explicit action landscape. The talk also highlights the universality of the methods, suggesting applicability to other exactly solvable kernels.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is high: the work builds on well-known results by Smoluchowski and Lushnikov, and uses established techniques from field theory and large deviation theory. The speaker is careful to acknowledge limitations (e.g., paper not yet on arXiv) and to cross-check with simulations. The title matches the content precisely, covering exact fluctuation theory, product-kernel aggregation, instantons, and non-convex large deviations. No discordant sources are mentioned; the only external reference is the program webpage. No comments were provided for analysis.

213 words

Title / Content Match

The title accurately reflects the content: the talk focuses on exact fluctuation theory for product-kernel aggregation, including instantons and non-convex large deviations.

Quality & Reliability

8/10

The talk presents original exact results in non-equilibrium statistical physics, with rigorous mathematical derivations and support from numerical simulations. The speaker is a recognized expert at a reputed institute, but the results are not yet peer-reviewed (the paper was not yet on arXiv).

Key Moments

Cited Sources

Contribution & Novelties

The talk provides the first exact solution for the complete large deviation function (including non-convex parts) in a many-body non-equilibrium system with interactions. It explicitly demonstrates ensemble inequivalence arising from non-equilibrium dynamics, and offers instanton trajectories for each rare fluctuation. These results are a significant advance in the field of macroscopic fluctuation theory and exactly solvable stochastic processes.

Pour aller plus loin :

107 words

Radar Profile

The radar profile shows very high scores in technical level and information quality, reflecting a dense, rigorous, and expert-level presentation. The information quantity is high, but the global reliability is slightly lower due to the unpublished nature of the results. The overall profile indicates a talk of exceptional depth intended for specialists.

Reliability 8/10