arXiv subjects

Christoph Raab

Publications and source records attributed to Christoph Raab.

11 recordsLinked to original sources

arXiv2025

Multi-Messenger Search for Neutrino and Gravitational-Wave Emissions from Binary Black Holes Near Active Galactic Nuclei

Binary black holes (BBHs) in the vicinity of Active Galactic Nuclei (AGNs) are particularly interesting systems from both a cosmological and astrophysical point of view. Matter and radiation fields within the dense AGN environment could produce electromagnetic and neutrino emission in addition to gravitational waves (GWs). Moreover, interactions between BBHs and AGN accretion disks are expected to influence BBH formation channels and merger rates. Understanding these sources could help explain the unexpectedly high BBH masses observed through GWs by the LIGO-Virgo-KAGRA collaborations. We present a search for coincident gravitational-wave and neutrino emission from AGNs. Our innovative approach combines information from gravitational-wave data, neutrino observations, and AGN optical catalogs to increase the chances of identifying potential sources and studying their properties. We assess the sensitivity of the search using subthreshold gravitational-wave candidates from LIGO-Virgo-KAGRA data, neutrino event candidates from public IceCube Neutrino Observatory data and AGN candidates from the Quaia catalog. A confident detection of such an event would mark a breakthrough in multi-messenger astronomy.

astro-ph.HE↗

arXiv2025

Domain Borders Are There to Be Crossed With Federated Few-Shot Adaptation

Federated Learning has emerged as a leading paradigm for decentralized, privacy-preserving learning, particularly relevant in the era of interconnected edge devices equipped with sensors. However, the practical implementation of Federated Learning faces three primary challenges: the need for human involvement in costly data labelling processes for target adaptation, covariate shift in client device data collection due to environmental factors affecting sensors, leading to discrepancies between source and target samples, and the impracticality of continuous or regular model updates in resource-constrained environments due to limited data transmission capabilities and technical constraints on channel availability and energy efficiency. To tackle these issues, we expand upon an efficient and scalable Federated Learning framework tailored for real-world client adaptation in industrial settings. This framework leverages a pre-trained source model comprising a deep backbone, an adaptation module, and a classifier running on a powerful server. By freezing the backbone and classifier during client adaptation on resource-constrained devices, we allow the domain adaptive linear layer to handle target domain adaptation, thus minimizing overall computational overhead. Furthermore, this setup, designated as FedAcross+, is extended to encompass the processing of streaming data, thereby rendering the solution suitable for non-stationary environments. Extensive experimental results demonstrate the effectiveness of FedAcross+ in achieving competitive adaptation on low-end client devices with limited target samples, successfully addressing the challenge of domain shift. Moreover, our framework accommodates sporadic model updates within resource-constrained environments, ensuring practical and seamless deployment.

cs.LG↗

arXiv2025

IceCube searches for GeV neutrino counterparts associated with high-energy starting events

The origin of the astrophysical neutrino flux observed by IceCube is largely unknown. To help decipher its astrophysical origin, we propose an IceCube analysis that conducts follow-up searches for GeV neutrinos associated with neutrino events above 60 TeV, which are known to have a high probability to be of astrophysical origin. It could not only identify a new component of the astrophysical neutrino flux, but also characterize how its spectrum extrapolates from GeV to PeV energies. This would in turn give valuable insights into the internal processes of neutrino sources. Astrophysical transients, such as collapsars, have been proposed as sources of time-correlated GeV- and high-energy neutrinos. Conducting this search in such short time scales allows for a substantial reduction in the dominant background rate for GeV neutrino candidate events. We introduce the statistical method and sensitivity of this search as well as dedicated data quality checks. None of the searches yield statistically significant results, and we present the first limits on GeV neutrino emission associated to VHE neutrino events at short time scales.

astro-ph.HE↗

arXiv2025

Low-latency neutrino follow-up combining diverse IceCube selections

Neutrino observations are crucial for multi-messenger astronomy, but currently limited by effective area and atmospheric background. However, while other telescopes must point their limited field of view, IceCube can perform full-sky realtime follow-up of astrophysical transient sources with high uptime. For this, IceCube uses its Fast Response Analysis (FRA), which can provide results from an unbinned maximum likelihood analysis within tens of minutes of an astrophysical transient. Besides manually selected transient candidates, it also routinely scans areas of the sky compatible with gravitational wave alerts from LIGO/Virgo/KAGRA and the IceCube event singlets most likely to originate from an astrophysical source. Currently the analysis uses TeV muon neutrino candidate events whose signature permits especially precise angular reconstruction, selected and reconstructed at the South Pole and transmitted with low-latency over satellite. Recently, different event selections are also being included in IceCube analyses. These efforts include the follow-up of gravitational wave events with GeV neutrinos detected by IceCube-DeepCore and the observation of the Galactic plane with cascade events produced by all neutrino flavors. If made available on a day-scale latency, these event selections can also be used in FRAs. Moreso, multiple event samples can be combined in a Fast Response Analysis that is sensitive to a broader energy range of a neutrino transient spectrum and ensures the inclusion of all neutrino flavors. We present the analysis method and technical aspects of such an extension of the existing framework. This includes a proposed new pipeline allowing the inclusion of the more computationally-intensive reconstruction methods used by the aforementioned event selections. The extension is validated using example analyses implemented in this framework. (abbreviated)

astro-ph.HE↗

arXiv2022

Federated Learning -- Methods, Applications and beyond

In recent years the applications of machine learning models have increased rapidly, due to the large amount of available data and technological progress.While some domains like web analysis can benefit from this with only minor restrictions, other fields like in medicine with patient data are strongerregulated. In particular \emph{data privacy} plays an important role as recently highlighted by the trustworthy AI initiative of the EU or general privacy regulations in legislation. Another major challenge is, that the required training \emph{data is} often \emph{distributed} in terms of features or samples and unavailable for classicalbatch learning approaches. In 2016 Google came up with a framework, called \emph{Federated Learning} to solve both of these problems. We provide a brief overview on existing Methods and Applications in the field of vertical and horizontal \emph{Federated Learning}, as well as \emph{Fderated Transfer Learning}.

cs.LG↗

arXiv2020

Transfer learning extensions for the probabilistic classification vector machine

Transfer learning is focused on the reuse of supervised learning models in a new context. Prominent applications can be found in robotics, image processing or web mining. In these fields, the learning scenarios are naturally changing but often remain related to each other motivating the reuse of existing supervised models. Current transfer learning models are neither sparse nor interpretable. Sparsity is very desirable if the methods have to be used in technically limited environments and interpretability is getting more critical due to privacy regulations. In this work, we propose two transfer learning extensions integrated into the sparse and interpretable probabilistic classification vector machine. They are compared to standard benchmarks in the field and show their relevance either by sparsity or performance improvements.

cs.LG↗

arXiv2020

Reactive Soft Prototype Computing for Concept Drift Streams

The amount of real-time communication between agents in an information system has increased rapidly since the beginning of the decade. This is because the use of these systems, e. g. social media, has become commonplace in today's society. This requires analytical algorithms to learn and predict this stream of information in real-time. The nature of these systems is non-static and can be explained, among other things, by the fast pace of trends. This creates an environment in which algorithms must recognize changes and adapt. Recent work shows vital research in the field, but mainly lack stable performance during model adaptation. In this work, a concept drift detection strategy followed by a prototype-based adaptation strategy is proposed. Validated through experimental results on a variety of typical non-static data, our solution provides stable and quick adjustments in times of change.

cs.LG↗

arXiv2020

Low-Rank Subspace Override for Unsupervised Domain Adaptation

Current supervised learning models cannot generalize well across domain boundaries, which is a known problem in many applications, such as robotics or visual classification. Domain adaptation methods are used to improve these generalization properties. However, these techniques suffer either from being restricted to a particular task, such as visual adaptation, require a lot of computational time and data, which is not always guaranteed, have complex parameterization, or expensive optimization procedures. In this work, we present an approach that requires only a well-chosen snapshot of data to find a single domain invariant subspace. The subspace is calculated in closed form and overrides domain structures, which makes it fast and stable in parameterization. By employing low-rank techniques, we emphasize on descriptive characteristics of data. The presented idea is evaluated on various domain adaptation tasks such as text and image classification against state of the art domain adaptation approaches and achieves remarkable performance across all tasks.

cs.LG↗

arXiv2018

Absolute Frequency Atlas from 915 nm to 985 nm based on Laser Absorption Spectroscopy of Iodine

This article reports on laser absorption spectroscopy of iodine gas between 915 nm and 985 nm. This wavelength range is scanned utilizing a narrow linewidth and mode-hop-free tunable diode-laser whose frequency is actively controlled using a calibrated wavelength meter. This allows us to provide an iodine atlas that contains almost 10,000 experimentally observed reference lines with an uncertainty of 50 MHz. For common lines, good agreement is found with a publication by Gerstenkorn et al. [S. Gerstenkorn, P. Luc, Atlas du spectre d'absorption de la molecule d'iode 14800-20000 cm$^{-1}$, Paris: Editions du Centre National de la Recherche Scientifique (CNRS), 1978]. The new rich dataset allows existing models of the iodine molecule to be refined and can serve as a reference for laser frequency calibration and stabilization.

physics.atom-ph↗

arXiv2014

Passive intrinsic-linewidth narrowing of ultraviolet extended-cavity diode laser by weak optical feedback

We present a simple method for narrowing the intrinsic Lorentzian linewidth of a commercial ultraviolet grating extended-cavity diode laser (TOPTICA DL Pro) using weak optical feedback from a long external cavity. We achieve a suppression in frequency noise spectral density of 20 dB measured at frequencies around 1 MHz, corresponding to the narrowing of the intrinsic Lorentzian linewidth from 200 kHz to 2 kHz. The system is suitable for experiments requiring a tunable ultraviolet laser with narrow linewidth and low high-frequency noise, such as precision spectroscopy, optical clocks, and quantum information science experiments.

physics.optics↗

arXiv2004

Forces between a single atom and its distant mirror image

An excited-state atom whose emitted light is back-reflected by a distant mirror can experience trapping forces, because the presence of the mirror modifies both the electromagnetic vacuum field and the atom's own radiation reaction field. We demonstrate this mechanical action using a single trapped barium ion. We observe the trapping conditions to be notably altered when the distant mirror is shifted by an optical wavelength. The well-localised barium ion enables the spatial dependence of the forces to be measured explicitly. The experiment has implications for quantum information processing and may be regarded as the most elementary optical tweezers.

quant-ph↗