arXiv subjects

Manuel Weber

Publications and source records attributed to Manuel Weber.

At least 19 recordsLinked to original sources

arXiv2026

Dissipation-induced bulk and boundary criticality in the Haldane chain

We study the Haldane chain coupled to a dissipative ohmic bath. Using large-scale quantum Monte Carlo simulations, we identify a second-order quantum phase transition into an antiferromagnetic state with spontaneously broken SO(3) symmetry that is governed by an interacting fixed point with dynamical exponent $z\approx 2$. We derive a generalized string order parameter which indicates that the symmetry-protected topological ground state of the Haldane chain is stable for weak dissipation and develops a nontrivial scaling dimension at criticality. In particular, its topological edge modes display nontrivial boundary criticality that is distinct from an equivalent transition out of a trivial state; for the latter, we perform an accurate $\epsilon$ expansion of a dissipative $\phi^4$ theory with ordinary boundary conditions. Our setup realizes a spin impurity in a critical yet nonconformal antiferromagnet and paves the way for continuously tunable boundary criticality controlled by dissipation.

cond-mat.str-el↗

arXiv2025

Competing Dirac masses in one dimension: Symmetry-enhanced pseudo-first-order transition and deconfined criticality

Emergent symmetries and slow crossover phenomena are central themes in quantum criticality and manifest themselves in the pseudocritical scaling experienced in the context of deconfined criticality. Here we discover its conceptual counterpart, i.e., a symmetry-enhanced pseudo-first-order transition. It emerges from a one-dimensional realization of deconfined criticality between charge- and bond-ordered states driven by competing Holstein and Su-Schrieffer-Heeger electron-phonon couplings, for which quantum fluctuations and thereby the nature of the transition can be tuned systematically via the phonon frequency $\omega_0$. In the classical limit $\omega_0 \to 0$, a low-energy Dirac theory predicts a direct first-order transition with emergent U(1) symmetry. Using exact quantum Monte Carlo simulations, we provide strong evidence for symmetry enhancement and even finite-size scaling on intermediate length scales but in the thermodynamic limit it turns into a narrow intermediate phase where both order parameters are finite, as chiral U(1) symmetry is weakly broken on the lattice. Including quantum lattice fluctuations diminishes the width of the intermediate phase, gradually restores the U(1) symmetry, and eventually tunes the system to a deconfined quantum critical point.

cond-mat.str-el↗

arXiv2025

PyViT-FUSE: A Foundation Model for Multi-Sensor Earth Observation Data

We propose PyViT-FUSE, a foundation model for earth observation data explicitly designed to handle multi-modal imagery by learning to fuse an arbitrary number of mixed-resolution input bands into a single representation through an attention mechanism. The learned patch tokens are further processed by a stack of vision transformers with a novel pyramidal structure. We train the model on a globally sampled dataset in a self-supervised manner, leveraging core concepts of the SwAV algorithm. We show the interpretability of the fusion mechanism by visualization of the attention scores and the models applicability to downstream tasks.

cs.CV↗

arXiv2025

Digital Guardians: Can GPT-4, Perspective API, and Moderation API reliably detect hate speech in reader comments of German online newspapers?

In recent years, toxic content and hate speech have become widespread phenomena on the internet. Moderators of online newspapers and forums are now required, partly due to legal regulations, to carefully review and, if necessary, delete reader comments. This is a labor-intensive process. Some providers of large language models already offer solutions for automated hate speech detection or the identification of toxic content. These include GPT-4o from OpenAI, Jigsaw's (Google) Perspective API, and OpenAI's Moderation API. Based on the selected German test dataset HOCON34k, which was specifically created for developing tools to detect hate speech in reader comments of online newspapers, these solutions are compared with each other and against the HOCON34k baseline. The test dataset contains 1,592 annotated text samples. For GPT-4o, three different promptings are used, employing a Zero-Shot, One-Shot, and Few-Shot approach. The results of the experiments demonstrate that GPT-4o outperforms both the Perspective API and the Moderation API, and exceeds the HOCON34k baseline by approximately 5 percentage points, as measured by a combined metric of MCC and F2-score.

cs.CL↗

arXiv2024

Intermediate bond-order-wave phase and nature of the order-to-order transition in the one-dimensional Hubbard-Holstein model

The Hubbard-Holstein model is one of the central models that describe the competition between electron-electron and electron-phonon interactions. In one dimension and at half-filling, the interplay between an electronic spin-density wave and a phonon-driven charge-density wave is considered to stabilize an intermediate Luther-Emery liquid. Here we show that, once the extended metallic regime disappears, a narrow bond-order-wave phase emerges. We use an exact directed-loop quantum Monte Carlo method for retarded interactions to simulate system sizes of several hundred sites, necessary to observe the weak signatures of this novel phase. Our results suggest a second-order quantum phase transition between the two charge orders that only turns first-order at strong coupling. Our findings are reminiscent of the extended Hubbard model, but they are driven by competing frequency dependencies of the same dynamically-screened Hubbard interaction.

cond-mat.str-el↗

arXiv2024

EnergyPlus Room Simulator

Research towards energy optimization in buildings heavily relies on building-related data such as measured indoor climate factors. While data collection is a labor- and cost-intensive task, simulations are a cheap alternative to generate datasets of arbitrary sizes, particularly useful for data-intensive deep learning methods. In this paper, we present the tool EnergyPlus Room Simulator, which enables the simulation of indoor climate in a specific room of a building using the simulation software EnergyPlus. It allows to alter room models and simulate various factors such as temperature, humidity, and CO2 concentration. In contrast to manually working with EnergyPlus, this tool enhances the simulation process by offering a convenient interface, including a user-friendly graphical user interface (GUI) as well as a REST API. The tool is intended to support scientific, building-related tasks such as occupancy detection on a room level by facilitating fast access to simulation data that may, for instance, be used for pre-training machine learning models.

cs.LG↗

arXiv2024

Unified Deep Learning Model for Global Prediction of Aboveground Biomass, Canopy Height and Cover from High-Resolution, Multi-Sensor Satellite Imagery

Regular measurement of carbon stock in the world's forests is critical for carbon accounting and reporting under national and international climate initiatives, and for scientific research, but has been largely limited in scalability and temporal resolution due to a lack of ground based assessments. Increasing efforts have been made to address these challenges by incorporating remotely sensed data. We present a new methodology which uses multi-sensor, multi-spectral imagery at a resolution of 10 meters and a deep learning based model which unifies the prediction of above ground biomass density (AGBD), canopy height (CH), canopy cover (CC) as well as uncertainty estimations for all three quantities. The model is trained on millions of globally sampled GEDI-L2/L4 measurements. We validate the capability of our model by deploying it over the entire globe for the year 2023 as well as annually from 2016 to 2023 over selected areas. The model achieves a mean absolute error for AGBD (CH, CC) of 26.1 Mg/ha (3.7 m, 9.9 %) and a root mean squared error of 50.6 Mg/ha (5.4 m, 15.8 %) on a globally sampled test dataset, demonstrating a significant improvement over previously published results. We also report the model performance against independently collected ground measurements published in the literature, which show a high degree of correlation across varying conditions. We further show that our pre-trained model facilitates seamless transferability to other GEDI variables due to its multi-head architecture.

cs.CV↗

arXiv2024

Tunable quantum criticality and pseudocriticality across the fixed-point annihilation in the anisotropic spin-boson model

Spin-boson models are simple examples of quantum dissipative systems, but also serve as effective models in quantum magnetism and exhibit nontrivial criticality. Recently, they have been established as a platform to study the nontrivial renormalization-group (RG) scenario of fixed-point annihilation, in which two intermediate-coupling RG fixed points collide and generate an extremely slow RG flow near the collision. For the Bose Kondo model, a single $S=1/2$ spin where each spin component couples to an independent bosonic bath with power-law spectrum $\propto \omega^s$ via dissipation strengths $\alpha_i$, $i\in\{x,y,z\}$, such phenomena occur sequentially for the U(1)-symmetric model at $\alpha_z=0$ and the SU(2)-symmetric case at $\alpha_z = \alpha_{xy}$, as the bath exponent $s<1$ is tuned. Here we use an exact wormhole quantum Monte Carlo method for retarded interactions to explore how this nontrivial fixed-point structure affects the phase diagram and phase transitions of the anisotropic model. In particular, we show how fixed-point annihilation within a symmetry-enhanced critical manifold leads to (i) a continuous order-to-order transition beyond the Landau paradigm, (ii) a symmetry-enhanced first-order transition, and (iii) pseudocriticality, which can be tuned into each other via the bath exponent $s$. We extract critical exponents at the continuous transition, but also find scaling behavior at the symmetry-enhanced first-order transition, for which the inverse correlation-length exponent is given by the bath exponent $s$. Moreover, we provide direct numerical evidence for pseudocritical scaling on both sides of the fixed-point collision, which manifests in an extremely slow drift of the correlation-length exponent. We also study the crossover away from the SU(2)-symmetric case and determine the phase boundary of an extended U(1)-symmetric critical phase.

cond-mat.str-el↗

arXiv2023

Dissipation-induced long-range order in the one-dimensional Bose-Hubbard model

Understanding the stability of strongly correlated phases of matter when coupled to environmental degrees of freedom is crucial for identifying the conditions under which these states may be observed. Here, we focus on the paradigmatic one-dimensional Bose-Hubbard model, and study the stability of the Luttinger liquid and Mott insulating phases in the presence of local particle exchange with site-independent baths of non-interacting bosons. We perform a numerically exact analysis of this model by adapting the recently developed wormhole quantum Monte Carlo method for retarded interactions to a continuous-time formulation with worm updates; we show how the wormhole updates can be easily implemented in this scheme. For an Ohmic bath, our numerical findings confirm the scaling prediction that the Luttinger-liquid phase becomes unstable at infinitesimal bath coupling. We show that the ensuing phase is a long-range ordered superfluid with spontaneously-broken U(1) symmetry. While the Mott insulator remains a distinct phase for small bath coupling, it exhibits diverging compressibility and non-integer local boson occupation in the thermodynamic limit. Upon increasing the bath coupling, this phase undergoes a transition to a long-range ordered superfluid. Finally, we discuss the effects of super-Ohmic dissipation on the Luttinger-liquid phase. Our results are compatible with a stable dissipative Luttinger-liquid phase that transitions to a long-range ordered superfluid at a finite system-bath coupling.

cond-mat.str-el↗

arXiv2023

Quantum spin chains with bond dissipation

We study the effect of bond dissipation on the one-dimensional antiferromagnetic spin-$1/2$ Heisenberg model. In analogy to the spin-Peierls problem, the dissipative bath is described by local harmonic oscillators that modulate the spin exchange coupling, but instead of a single boson frequency we consider a continuous bath spectrum $\propto \omega^s$. Using an exact quantum Monte Carlo method for retarded interactions, we show that for $s<1$ any finite coupling to the bath induces valence-bond-solid order, whereas for $s>1$ the critical phase of the isolated chain remains stable up to a finite critical coupling. We find that, even in the presence of the gapless bosonic spectrum, the spin-triplet gap remains well defined for any system size, from which we extract a dynamical critical exponent of $z=1$. We provide evidence for a Berezinskii-Kosterlitz-Thouless quantum phase transition that is governed by the SU(2)$_1$ Wess-Zumino-Witten model. Our results suggest that the critical properties of the dissipative system are the same as for the spin-Peierls model, irrespective of the different interaction range, i.e., power-law vs. exponential decay, of the retarded dimer-dimer interaction, indicating that the spin-Peierls criticality is robust with respect to the bosonic density of states.

cond-mat.str-el↗

arXiv2022

Dissipation-Induced Order: The $S=1/2$ Quantum Spin Chain Coupled to an Ohmic Bath

We consider an $S=1/2$ antiferromagnetic quantum Heisenberg chain where each site is coupled to an independent bosonic bath with ohmic dissipation. The coupling to the bath preserves the global SO(3) spin symmetry. Using large-scale, approximation-free quantum Monte Carlo simulations, we show that any finite coupling to the bath suffices to stabilize long-range antiferromagnetic order. This is in stark contrast to the isolated Heisenberg chain where spontaneous breaking of the SO(3) symmetry is forbidden by the Mermin-Wagner theorem. A linear spin-wave theory analysis confirms that the memory of the bath and the concomitant retarded interaction stabilize the order. For the Heisenberg chain, the ohmic bath is a marginal perturbation so that exponentially large system sizes are required to observe long-range order at small couplings. Below this length scale, our numerics is dominated by a crossover regime where spin correlations show different power-law behaviors in space and time. We discuss the experimental relevance of this crossover phenomena.

cond-mat.str-el↗

arXiv2022

Quantum Monte Carlo simulation of spin-boson models using wormhole updates

We present an exact quantum Monte Carlo method for spin systems coupled to dissipative bosonic baths which makes use of nonlocal wormhole updates to simulate the retarded spin-flip interactions originating from an off-diagonal spin-boson coupling. The method is closely related to the stochastic series expansion and extends the scope of the global directed-loop updates to nonlocal moves through a world-line configuration. We test our method for the U(1)-symmetric two-bath spin-boson model, where the off-diagonal components of a spin-$1/2$ particle are coupled to identical independent baths with power-law spectra, and get a precise estimate of the critical coupling between the critical and the localized phase. Our method applies to impurity systems and lattice models in any spatial dimension coupled to bosonic modes with arbitrary spectral distributions.

cond-mat.str-el↗

arXiv2022

Theoretical description of time-resolved photoemission in charge-density-wave materials out to long times

We describe coupled electron-phonon systems semiclassically - Ehrenfest dynamics for the phonons and quantum mechanics for the electrons - using a classical Monte Carlo approach that determines the nonequilibrium response to a large pump field. The semiclassical approach is quite accurate, because the phonons are excited to average energies much higher than the phonon frequency, eliminating the need for a quantum description. The numerical efficiency of this method allows us to perform a self-consistent time evolution out to very long times (tens of picoseconds) enabling us to model pump-probe experiments of a charge density wave (CDW) material. Our system is a half-filled, one-dimensional (1D) Holstein chain that exhibits CDW ordering due to a Peierls transition. The chain is subjected to a time-dependent electromagnetic pump field that excites it out of equilibrium, and then a second probe pulse is applied after a time delay. By evolving the system to long times, we capture the complete process of lattice excitation and subsequent relaxation to a new equilibrium, due to an exchange of energy between the electrons and the lattice, leading to lattice relaxation at finite temperatures. We employ an indirect (impulsive) driving mechanism of the lattice by the pump pulse due to the driving of the electrons by the pump field. We identify two driving regimes, where the pump can either cause small perturbations or completely invert the initial CDW order. Our work successfully describes the ringing of the amplitude mode in CDW systems that has long been seen in experiment, but never successfully explained by microscopic theory.

cond-mat.str-el↗

arXiv2022

SU(2)-Symmetric Spin-Boson Model: Quantum Criticality, Fixed-Point Annihilation, and Duality

The annihilation of two intermediate-coupling renormalization-group (RG) fixed points is of interest in diverse fields from statistical mechanics to high-energy physics, but has so far only been studied using perturbative techniques. Here we present high-accuracy quantum Monte Carlo results for the SU(2)-symmetric $S=1/2$ spin-boson (or Bose-Kondo) model. We study the model with a power-law bath spectrum $\propto \omega^s$ where, in addition to a critical phase predicted by perturbative RG, a stable strong-coupling phase is present. Using a detailed scaling analysis, we provide direct numerical evidence for the collision and annihilation of two RG fixed points at $s^\ast = 0.6540(2)$, causing the critical phase to disappear for $s<s^\ast$. In particular, we uncover a surprising duality between the two fixed points, corresponding to a reflection symmetry of the RG beta function, which we utilize to make analytical predictions at strong coupling which are in excellent agreement with numerics. Our work makes phenomena of fixed-point annihilation accessible to large-scale simulations, and we comment on the consequences for impurity moments in critical magnets.

cond-mat.str-el↗

arXiv2021

Real-time evolution of static electron-phonon models in time-dependent electric fields

We present an exact Monte Carlo method to simulate the nonequilibrium dynamics of electron-phonon models in the adiabatic limit of zero phonon frequency. The classical nature of the phonons allows us to sample the equilibrium phonon distribution and efficiently evolve the electronic subsystem in a time-dependent electromagnetic field for each phonon configuration. We demonstrate that our approach is particularly useful for charge-density-wave systems experiencing pulsed electric fields, as they appear in pump-probe experiments. For the half-filled Holstein model in one and two dimensions, we calculate the out-of-equilibrium response of the current and the energy after a pulse is applied as well as the photoemission spectrum before and after the pump. Finite-size effects are under control for chains of $162$ sites (in one dimension) or $16\times 16$ square lattices (in two dimensions).

cond-mat.str-el↗

arXiv2021

Electronic Mechanism that Quenches Field-Driven Heating as Illustrated with the Static Holstein Model

Time-dependent driving of quantum systems has emerged as a powerful tool to engineer exotic phases far from thermal equilibrium, but in the presence of many-body interactions it also leads to runaway heating, so that generic systems are believed to heat up until they reach a featureless infinite-temperature state. Understanding the mechanisms by which such a heat death can be slowed down or even avoided is a major goal -- one such mechanism is to drive toward an even distribution of electrons in momentum space. Here we show how such a mechanism avoids runaway heating for an interacting charge-density-wave chain with a macroscopic number of conserved quantities when driven by a strong dc electric field; minibands with nontrivial distribution functions develop as the current is prematurely driven to zero. Moreover, when approaching a zero-temperature resonance, the field strength can tune between positive, negative, or close-to-infinite effective temperatures for each miniband. Our results suggest that nontrivial metastable distribution functions should be realized in the prethermal regime of quantum systems coupled to slow bosonic modes.

cond-mat.str-el↗

arXiv2020

Towards the Detection of Building Occupancy with Synthetic Environmental Data

Information about room-level occupancy is crucial to many building-related tasks, such as building automation or energy performance simulation. Current occupancy detection literature focuses on data-driven methods, but is mostly based on small case studies with few rooms. The necessity to collect room-specific data for each room of interest impedes applicability of machine learning, especially data-intensive deep learning approaches, in practice. To derive accurate predictions from less data, we suggest knowledge transfer from synthetic data. In this paper, we conduct an experiment with data from a CO$_2$ sensor in an office room, and additional synthetic data obtained from a simulation. Our contribution includes (a) a simulation method for CO$_2$ dynamics under randomized occupant behavior, (b) a proof of concept for knowledge transfer from simulated CO$_2$ data, and (c) an outline of future research implications. From our results, we can conclude that the transfer approach can effectively reduce the required amount of data for model training.

cs.LG↗

arXiv2020

Valence-Bond Order in a Honeycomb Antiferromagnet Coupled to Quantum Phonons

We use exact quantum Monte Carlo simulations to demonstrate that the Néel ground state of an antiferromagnetic SU(2) spin-$\frac{1}{2}$ Heisenberg model on the honeycomb lattice can be destroyed by a coupling to quantum phonons. We find a clear first-order transition to a valence-bond-solid state with Kekulé order instead of a deconfined quantum critical point. However, quantum lattice fluctuations can drive the transition towards weakly first-order, revealing a tunability of the transition by the retardation of the interaction. In contrast to the one-dimensional case, our phase diagram in the adiabatic regime is qualitatively different from the frustrated $J_1$-$J_2$ model. Our results suggest that a coupling to bond phonons can induce Kekulé order in Dirac systems.

cond-mat.str-el↗