Anthony D. Kennedy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Anthony D. Kennedy has a total of 273 co-authors. Here are the top co-authors by number of shared publications:
- Urs M. Heller
American Physical Society 11961 Ridge New York USA
27 shared publications D. Toussaint
Department of Physics, University of Arizona, Tucson, Arizona 85721
26 shared publicationsSteven Gottlieb
Department of Physics, Indiana University, Bloomington, Indiana 47405
21 shared publicationsA. Krasnitz
IPS, RZ F3, Eidgenössische Technische Hochschule-Zentrum, CH-8092 Zurich, Switzerland
19 shared publications- Brian J. Pendleton
Department of Physics, The University of Edinburgh, Edinburgh EH9 3JZ, Scotland, UK
19 shared publications Pietro Rossi
Thinking Machines Corporation, Cambridge, Massachusetts 02139
19 shared publications
Latest Publications
- Open Access
- 21 July 2025
- Published by American Physical Society (APS) in Physical Review B
- Vol. 112 (4)
Fully ergodic simulations using radial updates
A sensible application of the hybrid Monte Carlo (HMC) method is often hindered by the presence of large—or even infinite—potential barriers. These potential barriers separate the configuration space into distinct sectors and can lead to ergodicity violations that bias measurements. In this work, we address this problem by augmenting HMC with a multiplicative Metropolis-Hastings update in a “radial direction” of the fields which enables crossing the potential barriers and ensures ergodicity of the sampling algorithm at comparably low computational cost. We demonstrate the algorithm on a simple toy model and show how it can be applied to the fermionic Hubbard model describing physics ranging from an exactly solvable two-site system to the C 20 H 12 perylene molecule. Our numerical results show that the radial updates successfully remove ergodicity violations, while simultaneously reducing autocorrelation times.
- 21 July 2025
- Published by American Physical Society (APS) in Physical Review B
- Vol. 112 (4)
Open AccessFully ergodic simulations using radial updates
A sensible application of the hybrid Monte Carlo (HMC) method is often hindered by the presence of large—or even infinite—potential barriers. These potential barriers separate the configuration space into distinct sectors and can lead to ergodicity violations that bias measurements. In this work, we address this problem by augmenting HMC with a multiplicative Metropolis-Hastings update in a “radial direction” of the fields which enables crossing the potential barriers and ensures ergodicity of the sampling algorithm at comparably low computational cost. We demonstrate the algorithm on a simple toy model and show how it can be applied to the fermionic Hubbard model describing physics ranging from an exactly solvable two-site system to the C 20 H 12 perylene molecule. Our numerical results show that the radial updates successfully remove ergodicity violations, while simultaneously reducing autocorrelation times.
- Open Access
- 17 February 2025
- Published by Sissa Medialab Srl
- Vol. 466, 064
On the geometric convergence of HMC on Riemannian manifolds
We show HMC is convergent in at least a geometric speed on any compact complete Riemannian manifold as state space; If otherwise the state space is non-compact and the potential diverges, HMCwithanextra Metropolis step along the Radial direction also converges geometrically. This work provides general results on conditions satisfied by component steps hence may be extended to convergence of other algorithms.
- 17 February 2025
- Published by Sissa Medialab Srl
- Vol. 466, 064
Open AccessOn the geometric convergence of HMC on Riemannian manifolds
We show HMC is convergent in at least a geometric speed on any compact complete Riemannian manifold as state space; If otherwise the state space is non-compact and the potential diverges, HMCwithanextra Metropolis step along the Radial direction also converges geometrically. This work provides general results on conditions satisfied by component steps hence may be extended to convergence of other algorithms.
- Open Access
- 5 December 2024
- Published by Sissa Medialab Srl
- Vol. 466, 068
Overcoming Ergodicity Problems of the Hybrid Monte Carlo Method using Radial Updates
Despite its many advantages, the sensible application of the Hybrid Monte Carlo (HMC) method is often hindered by the presence of large - or even infinite - potential barriers. These potential barriers partition the configuration space into distinct sectors, which leads to ergodicity violations and biased measurements of observables. In this work, we address this problem by augmenting the HMC method with a multiplicative Metropolis-Hastings update in a so-called "radial direction" of the fields, which enables jumps over the aforementioned potential barriers at comparably low computational cost. The effectiveness of this approach is demonstrated for the Hubbard model, formulated in a non-compact space by means of a continuous Hubbard-Stratonovich transformation. Our numerical results show that the radial updates successfully resolve the ergodicity violation, while simultaneously reducing autocorrelations.
- 5 December 2024
- Published by Sissa Medialab Srl
- Vol. 466, 068
Open AccessOvercoming Ergodicity Problems of the Hybrid Monte Carlo Method using Radial Updates
Despite its many advantages, the sensible application of the Hybrid Monte Carlo (HMC) method is often hindered by the presence of large - or even infinite - potential barriers. These potential barriers partition the configuration space into distinct sectors, which leads to ergodicity violations and biased measurements of observables. In this work, we address this problem by augmenting the HMC method with a multiplicative Metropolis-Hastings update in a so-called "radial direction" of the fields, which enables jumps over the aforementioned potential barriers at comparably low computational cost. The effectiveness of this approach is demonstrated for the Hubbard model, formulated in a non-compact space by means of a continuous Hubbard-Stratonovich transformation. Our numerical results show that the radial updates successfully resolve the ergodicity violation, while simultaneously reducing autocorrelations.
- Open Access
- 21 November 2024
- Published by Springer Nature in Journal of High Energy Physics
- Vol. 2024 (11) , 1-58
Tensor reduction for Feynman integrals with Lorentz and spinor indices
We present an efficient graphical approach to construct projectors for the tensor reduction of multi-loop Feynman integrals with both Lorentz and spinor indices in D dimensions. An ansatz for the projectors is constructed making use of its symmetry properties via an orbit partition formula. The graphical approach allows to identify and enumerate the orbits in each case. For the case without spinor indices we find a 1 to 1 correspondence between orbits and integer partitions describing the cycle structure of certain bi-chord graphs. This leads to compact combinatorial formulae for the projector ansatz. With spinor indices the graph-structure becomes more involved, but the method is equally applicable. Our spinor reduction formulae are based on the antisymmetric basis of γ matrices, and make use of their orthogonality property. We also provide a new compact formula to pass into the antisymmetric basis. We compute projectors for vacuum tensor Feynman integrals with up to 32 Lorentz indices and up to 4 spinor indices. We discuss how to employ the projectors in problems with external momenta.
- 21 November 2024
- Published by Springer Nature in Journal of High Energy Physics
- Vol. 2024 (11) , 1-58
Open AccessTensor reduction for Feynman integrals with Lorentz and spinor indices
We present an efficient graphical approach to construct projectors for the tensor reduction of multi-loop Feynman integrals with both Lorentz and spinor indices in D dimensions. An ansatz for the projectors is constructed making use of its symmetry properties via an orbit partition formula. The graphical approach allows to identify and enumerate the orbits in each case. For the case without spinor indices we find a 1 to 1 correspondence between orbits and integer partitions describing the cycle structure of certain bi-chord graphs. This leads to compact combinatorial formulae for the projector ansatz. With spinor indices the graph-structure becomes more involved, but the method is equally applicable. Our spinor reduction formulae are based on the antisymmetric basis of γ matrices, and make use of their orthogonality property. We also provide a new compact formula to pass into the antisymmetric basis. We compute projectors for vacuum tensor Feynman integrals with up to 32 Lorentz indices and up to 4 spinor indices. We discuss how to employ the projectors in problems with external momenta.
- Open Access
Overcoming Ergodicity Problems of the Hybrid Monte Carlo Method using Radial Updates
Despite its many advantages, the sensible application of the Hybrid Monte Carlo (HMC) method is often hindered by the presence of large - or even infinite - potential barriers. These potential barriers partition the configuration space into distinct sectors, which leads to ergodicity violations and biased measurements of observables. In this work, we address this problem by augmenting the HMC method with a multiplicative Metropolis-Hastings update in a so-called "radial direction" of the fields, which enables jumps over the aforementioned potential barriers at comparably low computational cost. The effectiveness of this approach is demonstrated for the Hubbard model, formulated in a non-compact space by means of a continuous Hubbard-Stratonovich transformation. Our numerical results show that the radial updates successfully resolve the ergodicity violation, while simultaneously reducing autocorrelations.
Open AccessOvercoming Ergodicity Problems of the Hybrid Monte Carlo Method using Radial Updates
Despite its many advantages, the sensible application of the Hybrid Monte Carlo (HMC) method is often hindered by the presence of large - or even infinite - potential barriers. These potential barriers partition the configuration space into distinct sectors, which leads to ergodicity violations and biased measurements of observables. In this work, we address this problem by augmenting the HMC method with a multiplicative Metropolis-Hastings update in a so-called "radial direction" of the fields, which enables jumps over the aforementioned potential barriers at comparably low computational cost. The effectiveness of this approach is demonstrated for the Hubbard model, formulated in a non-compact space by means of a continuous Hubbard-Stratonovich transformation. Our numerical results show that the radial updates successfully resolve the ergodicity violation, while simultaneously reducing autocorrelations.
- Open Access
Hamiltonian Monte Carlo on Symmetric and Homogeneous Spaces via Symplectic Reduction
The Hamiltonian Monte Carlo method generates samples by introducing a mechanical system that explores the target density. For distributions on manifolds it is not always simple to perform the mechanics as a result of the lack of global coordinates, the constraints of the manifold, and the requirement to compute the geodesic flow. In this paper we explain how to construct the Hamiltonian system on naturally reductive homogeneous spaces using symplectic reduction, which lifts the HMC scheme to a matrix Lie group with global coordinates and constant metric. This provides a general framework that is applicable to many manifolds that arise in applications, such as hyperspheres, hyperbolic spaces, symmetric positive-definite matrices, Grassmannian, and Stiefel manifolds.
Open AccessHamiltonian Monte Carlo on Symmetric and Homogeneous Spaces via Symplectic Reduction
The Hamiltonian Monte Carlo method generates samples by introducing a mechanical system that explores the target density. For distributions on manifolds it is not always simple to perform the mechanics as a result of the lack of global coordinates, the constraints of the manifold, and the requirement to compute the geodesic flow. In this paper we explain how to construct the Hamiltonian system on naturally reductive homogeneous spaces using symplectic reduction, which lifts the HMC scheme to a matrix Lie group with global coordinates and constant metric. This provides a general framework that is applicable to many manifolds that arise in applications, such as hyperspheres, hyperbolic spaces, symmetric positive-definite matrices, Grassmannian, and Stiefel manifolds.
- Open Access
- 7 March 2018
- Published by Annual Reviews in Annual Review of Statistics and Its Application
- Vol. 5 (1) , 451-471
Geometry and Dynamics for Markov Chain Monte Carlo
Markov chain Monte Carlo methods have revolutionized mathematical computation and enabled statistical inference within many previously intractable models. In this context, Hamiltonian dynamics have been proposed as an efficient way of building chains that can explore probability densities efficiently. The method emerges from physics and geometry, and these links have been extensively studied over the past thirty years. The aim of this review is to provide a comprehensive introduction to the geometric tools used in Hamiltonian Monte Carlo at a level accessible to statisticians, machine learners, and other users of the methodology with only a basic understanding of Monte Carlo methods. This will be complemented with some discussion of the most recent advances in the field, which we believe will become increasingly relevant to scientists.
- 7 March 2018
- Published by Annual Reviews in Annual Review of Statistics and Its Application
- Vol. 5 (1) , 451-471
Open AccessGeometry and Dynamics for Markov Chain Monte Carlo
Markov chain Monte Carlo methods have revolutionized mathematical computation and enabled statistical inference within many previously intractable models. In this context, Hamiltonian dynamics have been proposed as an efficient way of building chains that can explore probability densities efficiently. The method emerges from physics and geometry, and these links have been extensively studied over the past thirty years. The aim of this review is to provide a comprehensive introduction to the geometric tools used in Hamiltonian Monte Carlo at a level accessible to statisticians, machine learners, and other users of the methodology with only a basic understanding of Monte Carlo methods. This will be complemented with some discussion of the most recent advances in the field, which we believe will become increasingly relevant to scientists.
- Open Access
- 15 July 2016
- Published by Sissa Medialab Srl
- Vol. 251, 274
Renormalizability of the Schrödinger Functional.
- 15 July 2016
- Published by Sissa Medialab Srl
- Vol. 251, 274
Open AccessRenormalizability of the Schrödinger Functional.
- Open Access
Numerical Stochastic Perturbation Theory and Gradient Flow in φ^4 Theory
In this contribution we present an exploratory study of several novel methods for numerical stochastic perturbation theory. For the investigation we consider observables defined through the gradient flow in the simple {\phi}^4 theory.
Open AccessNumerical Stochastic Perturbation Theory and Gradient Flow in φ^4 Theory
In this contribution we present an exploratory study of several novel methods for numerical stochastic perturbation theory. For the investigation we consider observables defined through the gradient flow in the simple {\phi}^4 theory.
- Open Access
Renormalizability of the Schrödinger Functional
Symanzik showed that quantum field theory can be formulated on a space with boundaries by including suitable surface interactions in the action to implement boundary conditions. We show that to all orders in perturbation theory all the divergences induced by these surface interactions can be absorbed by a renormalization of their coefficients.
Open AccessRenormalizability of the Schrödinger Functional
Symanzik showed that quantum field theory can be formulated on a space with boundaries by including suitable surface interactions in the action to implement boundary conditions. We show that to all orders in perturbation theory all the divergences induced by these surface interactions can be absorbed by a renormalization of their coefficients.