Publications

Order by: publication date number of citations

• AdvNF: Reducing mode collapse in conditional normalising flows using adversarial learning

  Vikas Kanaujia, Mathias S. Scheurer, Vipul Arora

  SciPost Phys. 16, 132 (2024) · published 24 May 2024 | Toggle abstract · pdf

  Deep generative models complement Markov-chain-Monte-Carlo methods for efficiently sampling from high-dimensional distributions. Among these methods, explicit generators, such as Normalising Flows (NFs), in combination with the Metropolis Hastings algorithm have been extensively applied to get unbiased samples from target distributions. We systematically study central problems in conditional NFs, such as high variance, mode collapse and data efficiency. We propose adversarial training for NFs to ameliorate these problems. Experiments are conducted with low-dimensional synthetic datasets and XY spin models in two spatial dimensions.

• The acceptance speech of the 2022 Weyl-Wigner award

  Nicolai Reshetikhin

  SciPost Phys. Proc. 14, 003 (2023) · published 23 November 2023 | Toggle abstract · pdf

  This is the acceptance speech for the 2022 Weyl-Wigner award.

• The importance of multiple scatterings in medium-induced gluon radiation

  Carlota Andres, Fabio Dominguez, Marcos Gonzalez Martinez

  SciPost Phys. Proc. 10, 016 (2022) · published 10 August 2022 | Toggle abstract · pdf

  In this work we disentangle the underlying physical picture of the in-medium gluon radiation process across its different energy regimes by comparing the recently obtained fully-resummed -- without any further approximations -- BDMPS-Z in-medium emission spectrum with the extensively used analytical approaches. We observe that in the high-energy regime the radiation process is dominated by a single hard scattering, while in the intermediate-energy region coherence effects among multiple scatterings are crucial. Finally, we prove that in the low-energy regime the dynamics is again controlled by a single scattering but where one must include a suppression factor accounting for the probability of not having any further scatterings.

• A solvable model for graph state decoherence dynamics

  Jérôme Houdayer, Haggai Landa, Grégoire Misguich

  SciPost Phys. Core 7, 009 (2024) · published 23 February 2024 | Toggle abstract · pdf

  We present an exactly solvable toy model for the continuous dissipative dynamics of permutation-invariant graph states of $N$ qubits. Such states are locally equivalent to an $N$-qubit Greenberger-Horne-Zeilinger (GHZ) state, a fundamental resource in many quantum information processing setups. We focus on the time evolution of the state governed by a Lindblad master equation with the three standard single-qubit jump operators, the Hamiltonian part being set to zero. Deriving analytic expressions for the expectation values of observables expanded in the Pauli basis at all times, we analyze the nontrivial intermediate-time dynamics. Using a numerical solver based on matrix product operators, we simulate the time evolution for systems with up to 64 qubits and verify a numerically exact agreement with the analytical results. We find that the evolution of the operator space entanglement entropy of a bipartition of the system manifests a plateau whose duration increases logarithmically with the number of qubits, whereas all Pauli-operator products have expectation values decaying at most in constant time.

• $\mathcal{O}(α_s^3)$ corrections to semileptonic $b \to c$ decays in the heavy daughter approximation

  Kay Schoenwald

  SciPost Phys. Proc. 7, 027 (2022) · published 21 June 2022 | Toggle abstract · pdf

  We present our recent calculation of the third order corrections to the semileptonic $b \to c$ and the muon decays. The calculation has been performed in an expansion around the limit $m_c \sim m_b$, but shows decent convergence even for $m_c=0$ from which the contribution to the muon decay can be extracted. For the semileptonic $b \to c$ decay we find large perturbative corrections in the on-shell scheme which can be significantly reduced by changing to the kinetic scheme for the bottom quark mass. These results are important input for the inclusive determination of $|V_{cb}|$ and the Fermi coupling constant $G_F$.

• Scalable imaginary time evolution with neural network quantum states

  Eimantas Ledinauskas, Egidijus Anisimovas

  SciPost Phys. 15, 229 (2023) · published 6 December 2023 | Toggle abstract · pdf

  The representation of a quantum wave function as a neural network quantum state (NQS) provides a powerful variational ansatz for finding the ground states of many-body quantum systems. Nevertheless, due to the complex variational landscape, traditional methods often employ the computation of quantum geometric tensor, consequently complicating optimization techniques. Contributing to efforts aiming to formulate alternative methods, we introduce an approach that bypasses the computation of the metric tensor and instead relies exclusively on first-order gradient descent with Euclidean metric. This allows for the application of larger neural networks and the use of more standard optimization methods from other machine learning domains. Our approach leverages the principle of imaginary time evolution by constructing a target wave function derived from the Schrödinger equation, and then training the neural network to approximate this target. We make this method adaptive and stable by determining the optimal time step and keeping the target fixed until the energy of the NQS decreases. We demonstrate the benefits of our scheme via numerical experiments with 2D $J_1-J_2$, Heisenberg model, which showcase enhanced stability and energy accuracy in comparison to direct energy loss minimization. Importantly, our approach displays competitiveness with the well-established density matrix renormalization group method and NQS optimization with stochastic reconfiguration.

• Investigation of the background in coherent $J/\psi$ production at the EIC

  Wan Chang

  SciPost Phys. Proc. 8, 053 (2022) · published 12 July 2022 | Toggle abstract · pdf

  Understanding fundamental properties of nucleons and nuclei are among the most important scientific goals of the next-generation machine, the Electron-Ion Collider (EIC). With the unprecedented versatility provided by the EIC, it will provide answers to many standing puzzles and open questions in modern nuclear physics. One of the golden measurements at the EIC is coherent vector meson production in electron-nucleus (eA) scattering in order to obtain the spatial gluon density distribution in heavy nuclei. This requires the experiment to overcome an overwhelmingly large background arising from the incoherent diffractive production, where the nucleus mostly breaks up into fragments of particles in the far-forward direction close to the hadron beam rapidity. In this talk, we systematically study the rejection of incoherent $J/\psi$ production by vetoing products from the nuclear breakup - protons, neutrons, and photons, which is modeled with the BeAGLE event generator and the most up-to-date EIC Far-forward Interaction Region design.

• Exclusion process subject to conformational size changes

  Yvan Rousset, Luca Ciandrini, Norbert Kern

  SciPost Phys. 6, 077 (2019) · published 26 June 2019 | Toggle abstract · pdf

  We introduce a model for stochastic transport on a one-dimensional substrate with particles assuming different conformations during their stepping cycles. These conformations correspond to different footprints on the substrate: in order to advance, particles are subject to successive contraction and expansion steps with different characteristic rates. We thus extend the paradigmatic exclusion process, provide predictions for all regimes of these rates that are in excellent agreement with simulations, and show that the current-density relation may be affected considerably. Symmetries are discussed, and exploited. We discuss our results in the context of molecular motors, confronting a hand-over-hand and an inchworm stepping mechanism, as well as for ribosomes.

• The droplet formation-dissolution transition in different ensembles: Finite-size scaling from two perspectives

  Franz Paul Spitzner, Johannes Zierenberg, Wolfhard Janke

  SciPost Phys. 5, 062 (2018) · published 13 December 2018 | Toggle abstract · pdf

  The formation and dissolution of a droplet is an important mechanism related to various nucleation phenomena. Here, we address the droplet formation-dissolution transition in a two-dimensional Lennard-Jones gas to demonstrate a consistent finite-size scaling approach from two perspectives using orthogonal control parameters. For the canonical ensemble, this means that we fix the temperature while varying the density and vice versa. Using specialised parallel multicanonical methods for both cases, we confirm analytical predictions at fixed temperature (rigorously only proven for lattice systems) and corresponding scaling predictions from expansions at fixed density. Importantly, our methodological approach provides us with reference quantities from the grand canonical ensemble that enter the analytical predictions. Our orthogonal finite-size scaling setup can be exploited for theoretical and experimental investigations of general nucleation phenomena - if one identifies the corresponding reference ensemble and adapts the theory accordingly. In this case, our numerical approach can be readily translated to the corresponding ensembles and thereby proves very useful for numerical studies of equilibrium droplet formation, in general.

• Flavor deformations and supersymmetry enhancement in $4d$ $\mathcal{N}=2$ theories

  Usman Naseer, Charles Thull

  SciPost Phys. 13, 058 (2022) · published 21 September 2022 | Toggle abstract · pdf

  We study $\mathcal{N}=2$ theories on four-dimensional manifolds that admit a Killing vector $v$ with isolated fixed points. It is possible to deform these theories by coupling position-dependent background fields to the flavor current multiplet. The partition function of the deformed theory only depends on the value of the background scalar fields at the fixed points of $v$. For a single adjoint hypermultiplet, the partition function becomes independent of the supergravity as well as the flavor background if the scalars attain special values at the fixed points. For these special values, supersymmetry at the fixed points enhances from the Donaldson-Witten twist to the Marcus twist or the Vafa-Witten twist of $\mathcal{N}=4$ SYM. Our results explain the recently observed squashing independence of $\mathcal{N}=2^*$ theory on the squashed sphere and provide a new squashing independent point. Interpreted through the AGT-correspondence, this implies the $b$-independence of torus one-point functions of certain local operators in Liouville/Toda CFT. The position-dependent deformations imply relations between correlators of partially integrated operators in any $\mathcal{N}=2$ SCFT with flavor symmetries.