English
Related papers

Related papers: Homogenized Transformers

200 papers

We develop a rigorous treatment of discontinuous stochastic unitary evolution for a system of quantum particles that interacts singularly with quantum "bubbles" at random instants of time. This model of a "cloud chamber" allows to watch and…

Mathematical Physics · Physics 2007-05-23 V. P. Belavkin

A stochastic treatment yielding to the derivation of a general Fokker-Planck equation is presented to model the slow convergence towards equilibrium of mean-field systems due to finite-N effects. The thermalization process involves notably…

Mathematical Physics · Physics 2011-09-28 W. Ettoumi , M. -C. Firpo

The starting point for this work is a static macroscopic model for a high-contrast layered material in single-slip finite crystal plasticity, identified in [Christowiak & Kreisbeck, Calc. Var. PDE (2017)] as a homogenization limit via…

Analysis of PDEs · Mathematics 2021-08-03 Elisa Davoli , Carolin Kreisbeck

Homology decomposition techniques are a powerful tool used in the analysis of the homotopy theory of (classifying) spaces. The associated Bousfield-Kan spectral sequences involve higher derived limits of the inverse limit functor. We study…

Algebraic Topology · Mathematics 2009-05-29 Dietrich Notbohm

The thermodynamics and the dynamics of particle systems with infinite-range coupling display several unusual and new features with respect to systems with short-range interactions. The Hamiltonian Mean Field (HMF) model represents a…

Statistical Mechanics · Physics 2009-09-29 Thierry Dauxois , Vito Latora , Andrea Rapisarda , Stefano Ruffo , Alessandro Torcini

Transformers have reshaped machine learning by utilizing attention mechanisms to capture complex patterns in large datasets, leading to significant improvements in performance. This success has contributed to the belief that "bigger means…

Machine Learning · Computer Science 2025-05-28 Hemanth Saratchandran , Damien Teney , Simon Lucey

The dynamics of a degree of freedom associated to an axial vector in contact with a heat bath is decribed by means of a probability distribution function obeying a Fokker-Planck equation. The equation is derived by using mesoscopic…

Soft Condensed Matter · Physics 2010-01-27 J. G. Méndez-Bermúdez , I. Santamaría-Holek

Transformers are extremely successful machine learning models whose mathematical properties remain poorly understood. Here, we rigorously characterize the behavior of transformers with hardmax self-attention and normalization sublayers as…

Computation and Language · Computer Science 2026-05-14 Albert Alcalde , Giovanni Fantuzzi , Enrique Zuazua

In the mean field integrate-and-fire model, the dynamics of a typical neuron within a large network is modeled as a diffusion-jump stochastic process whose jump takes place once the voltage reaches a threshold. In this work, the main goal…

Probability · Mathematics 2021-02-19 Jian-Guo Liu , Ziheng Wang , Yantong Xie , Yuan Zhang , Zhennan Zhou

In the mean field integrate-and-fire model, the dynamics of a typical neuron within a large network is modeled as a diffusion-jump stochastic process whose jump takes place once the voltage reaches a threshold. In this work, the main goal…

Probability · Mathematics 2023-06-22 Jian-Guo Liu , Ziheng Wang , Yantong Xie , Yuan Zhang , Zhennan Zhou

Although coarse-grained models have been widely used to explain exotic phenomena in complex fluids, such as droplet formation in living cells, these conventional approaches often fail to capture the intricate microscopic degrees of freedom…

Soft Condensed Matter · Physics 2025-06-13 Masanari Shimada , Tetsuya J. Kobayashi

With recently developed tools, we prove a homogenisation theorem for a random ODE with short and long-range dependent fractional noise. The effective dynamics are not necessarily diffusions, they are given by stochastic differential…

Probability · Mathematics 2020-06-23 Johann Gehringer , Xue-Mei Li

We consider a diffusion process with coefficients that are periodic outside of an 'interface region' of finite thickness. The question investigated in the articles [1,2] is the limiting long time / large scale behaviour of such a process…

Probability · Mathematics 2009-10-05 Martin Hairer , Charles Manson

We use dynamic light scattering and computer simulations to study equilibrium dynamics and dynamic heterogeneity in concentrated suspensions of colloidal hard spheres. Our study covers an unprecedented density range and spans seven decades…

Despite the Transformer's dominance across machine learning, its architecture remains largely heuristic and lacks a unified theoretical foundation. We introduce Score-based Variational Flow (SVFlow), a continuous-time dynamical system for…

Machine Learning · Computer Science 2026-04-28 Huadong Liao

We study closed dense collections of hard spheres that collide inelastically with constant coefficient of normal restitution. We find inhomogeneous states (IS) where the density profile is spatially non-uniform but constant in time. The…

Chaotic Dynamics · Physics 2014-02-19 Itzhak Fouxon

The homogeneous Boltzmann equation for inelastic Maxwell mixtures is considered to study the dynamics of tracer particles or impurities (solvent) immersed in a uniform granular gas (solute). The analysis is based on exact results derived…

Statistical Mechanics · Physics 2013-01-21 V. Garzó , E. Trizac

This article is concerned with the fluctuations and the concentration properties of a general class of discrete generation and mean field particle interpretations of nonlinear measure valued processes. We combine an original stochastic…

Probability · Mathematics 2012-11-09 Pierre Del Moral , Emmanuel Rio

Highly concentrated patterns have been observed in a spatially heterogeneous, nonlocal, model of BGK type implementing a velocity-jump process. We study both a linear and a nonlinear case and describe the concentration profile. In…

Mathematical Physics · Physics 2024-01-31 Nadia Loy , Benoit Perthame

We model the evolution of tokens within a deep stack of Transformer layers as a continuous-time flow on the unit sphere, governed by a mean-field interacting particle system, building on the framework introduced in (Geshkovski et al.,…

Machine Learning · Computer Science 2025-04-17 Giuseppe Bruno , Federico Pasqualotto , Andrea Agazzi