arXiv subjects

Jan Kessler

Publications and source records attributed to Jan Kessler.

6 recordsLinked to original sources

arXiv2026

Separable Force Matching of PBE0 Hybrid-Functional Reference Forces for Path-Integral Simulations of Liquid Water

Force-matched water models provide a practical route from first-principles reference data to long classical and path-integral molecular simulations. Previous flexible four-site potentials in the spirit of q-TIP4P/F showed that fitting analytic models to density-functional-theory forces can reproduce key structural features of liquid water while retaining the efficiency required for quantum-nuclear sampling. Here we introduce two refinements aimed at making this strategy more accurate and more reproducible for molecular simulation. First, the production fit is based on PBE0 hybrid-functional reference forces and therefore includes the Hartree--Fock exact-exchange contribution in the electronic-structure target. Second, the parametrization is formulated as a separable nonlinear least-squares problem in which all linear force-field amplitudes are eliminated analytically for every trial set of nonlinear shape parameters. The resulting force-matching protocol lowers the dimension of the nonlinear search, reduces compensation between heterogeneous parameters, and enables a controlled comparison of Lennard-Jones and Buckingham oxygen--oxygen repulsion terms. Applied to CP2K reference forces for liquid water, the projected optimization reproduces target force distributions and yields radial distribution functions close to first-principles and neutron-scattering benchmarks. The Buckingham representation gives a more flexible short-range repulsive wall than a purely Lennard-Jones form, and the final flexible model remains stable in path-integral simulations. The method provides a transparent workflow for deriving simulation-ready water potentials from accurate hybrid density-functional reference forces.

physics.chem-ph↗

arXiv2026

Basis-free neural-network geminal and Jastrow factors for variational Monte Carlo

Neural-network quantum states offer a flexible route to compact many-electron wave functions, but their practical accuracy depends strongly on how fermionic antisymmetry, electron correlation, and optimization noise are treated. Here we combine an antisymmetrized geminal power (AGP) determinant with feed-forward neural networks that replace conventional basis-set expansions in the geminal and in two Jastrow-factor constructions. The resulting basis-free Jastrow--AGP ansatz is optimized by variational Monte Carlo and is designed to separate two tasks: the AGP part defines the nodal surface, while the neural-network Jastrow factor recovers dynamical correlation at fixed nodes. This separation makes it possible to distinguish errors associated with dynamical correlation from those caused by static, multireference correlation. Applications to the hydrogen molecule and the rectangular hydrogen tetramer show sub-millihartree accuracy when the AGP nodes are adequate, and expose the residual nodal limitation near the large-radius square geometry of the hydrogen tetramer. These results clarify where neural-network building blocks can improve a compact geminal ansatz and where additional nodal flexibility is required.

physics.comp-ph↗

arXiv2020

Dynamics of Cycles in Polyhedra I: The Isolation Lemma

A cycle $C$ of a graph $G$ is \emph{isolating} if every component of $G-V(C)$ is a single vertex. We show that isolating cycles in polyhedral graphs can be extended to larger ones: every isolating cycle $C$ of length $6 \leq |E(C)| < \left \lfloor \frac{2}{3}(|V(G)|+4) \right \rfloor$ implies an isolating cycle $C'$ of larger length that contains $V(C)$. By "hopping" iteratively to such larger cycles, we obtain a powerful and very general inductive motor for proving long cycles and computing them (we will give an algorithm with quadratic running time). This is the first step towards the so far elusive quest of finding a universal induction that captures longest cycles of polyhedral graph classes. Our motor provides also a method to prove linear lower bounds on the length of Tutte cycles, as $C'$ will be a Tutte cycle of $G$ if $C$ is. We prove in addition that $|E(C')| \leq |E(C)|+3$ if $G$ contains no face of size five, which gives a new tool for results about cycle spectra, and provides evidence that faces of size five may obstruct long cycles in many graph classes. We test our motor on the following conjecture about essentially 4-connected graphs. A planar graph is \emph{essentially $4$-connected} if it is 3-connected and every of its 3-separators is the neighborhood of a single vertex. Jackson and Wormald proved that every essentially 4-connected planar graph $G$ on $n$ vertices contains a cycle of length at least $\frac{2}{5}(n+2)$, and this result has recently been improved multiple times, culminating in the lower bound $\frac{5}{8}(n+2)$. However, the currently best known upper bound is given by an infinite family of such graphs in which no graph $G$ contains a cycle that is longer than $\left \lfloor \frac{2}{3}(n+4) \right \rfloor$; this upper bound is still unmatched. Using isolating cycles, we improve the lower bound to match the upper. All our results are tight.

cs.DS↗

arXiv2019

Artificial Neural Networks as Trial Wave Functions for Quantum Monte Carlo

Inspired by the universal approximation theorem and widespread adoption of artificial neural network techniques in a diversity of fields, we propose feed-forward neural networks as a general purpose trial wave function for quantum Monte Carlo simulations of continous many-body systems. Whereas for simple model systems the whole many-body wave function can be represented by a neural network, the antisymmetry condition of non-trivial fermionic systems is incorporated by means of a Slater determinant. To demonstrate the accuracy of our trial wave functions, we have studied an exactly solvable model system of two trapped interacting particles, as well as the hydrogen dimer.

physics.comp-ph↗

arXiv2018

i-PI 2.0: A Universal Force Engine for Advanced Molecular Simulations

Progress in the atomic-scale modelling of matter over the past decade has been tremendous. This progress has been brought about by improvements in methods for evaluating interatomic forces that work by either solving the electronic structure problem explicitly, or by computing accurate approximations of the solution and by the development of techniques that use the Born-Oppenheimer (BO) forces to move the atoms on the BO potential energy surface. As a consequence of these developments it is now possible to identify stable or metastable states, to sample configurations consistent with the appropriate thermodynamic ensemble, and to estimate the kinetics of reactions and phase transitions. All too often, however, progress is slowed down by the bottleneck associated with implementing new optimization algorithms and/or sampling techniques into the many existing electronic-structure and empirical-potential codes. To address this problem, we are thus releasing a new version of the i-PI software. This piece of software is an easily extensible framework for implementing advanced atomistic simulation techniques using interatomic potentials and forces calculated by an external driver code. While the original version of the code was developed with a focus on path integral molecular dynamics techniques, this second release of i-PI not only includes several new advanced path integral methods, but also offers other classes of algorithms. In other words, i-PI is moving towards becoming a universal force engine that is both modular and tightly coupled to the driver codes that evaluate the potential energy surface and its derivatives.

physics.chem-ph↗

arXiv2015

Structure and Dynamics of the Instantaneous Water/Vapor Interface Revisited by Path-Integral and Ab-Initio Molecular Dynamics Simulations

The structure and dynamics of the water/vapor interface is revisited by means of path-integral and second-generation Car-Parrinello ab-initio molecular dynamics simulations in conjunction with an instantaneous surface definition [A. P. Willard and D. Chandler, J. Phys. Chem. B 114, 1954 (2010)]. In agreement with previous studies, we find that one of the OH bonds of the water molecules in the topmost layer is pointing out of the water into the vapor phase, while the orientation of the underlying layer is reversed. Therebetween, an additional water layer is detected, where the molecules are aligned parallel to the instantaneous water surface.

physics.chem-ph↗