Jakob Spiegelberg

A central question in machine learning is how reliable the predictions of a trained model are. Reliability includes the identification of instances for which a model is likely not to be trusted based on an analysis of the learning system itself. Such unreliability for an input may arise from the model family providing a variety of hypotheses consis...

The double descent phenomenon challenges traditional statistical learning theory by revealing scenarios where larger models do not necessarily lead to reduced performance on unseen data. While this counterintuitive behavior has been observed in a variety of classical machine learning models, particularly modern neural network architectures, it rema...

As quantum machine learning continues to develop at a rapid pace, the importance of ensuring the robustness and efficiency of quantum algorithms cannot be overstated. Our research presents an analysis of quantum randomized smoothing, how data encoding and perturbation modeling approaches can be matched to achieve meaningful robustness certificates....

Large Vision Language Models (LVLMs) have shown remarkable capabilities in multimodal tasks like visual question answering or image captioning. However, inconsistencies between the visual information and the generated text, a phenomenon referred to as hallucinations, remain an unsolved problem with regard to the trustworthiness of LVLMs. To address...

Quantifying predictive uncertainty of deep semantic segmentation networks is essential in safety-critical tasks. In applications like autonomous driving, where video data is available, convolutional long short-term memory networks are capable of not only providing semantic segmentations but also predicting the segmentations of the next timesteps. T...

We propose a magnetic measurement method utilizing a patterned postsample aperture in a transmission electron microscope. While utilizing electron magnetic circular dichroism, the method circumvents previous needs to shape the electron probe to an electron vortex beam or astigmatic beam. The method can be implemented in standard scanning transmissi...

Electron magnetic circular dichroism (EMCD), the electron wave analog of x-ray magnetic circular dichroism (XMCD), allows for the element specific measurement of the spin and orbital magnetic moments with up to nanometer resolution. However, due to dynamical diffraction effects, the signal-to-noise ratios of EMCD spectra are often very low. We desc...

We propose a magnetic measurement method utilizing a patterned post-sample aperture in a transmission electron microscope. While utilizing electron magnetic circular dichroism, the method circumvents previous needs to shape the electron probe to an electron vortex beam or astigmatic beam. The method can be implemented in standard scanning transmiss...

The complex interplay of elastic and inelastic scattering amenable to different levels of approximation constitutes the major challenge for the computation and hence interpretation of TEM-based spectroscopical methods. The two major approaches to calculate inelastic scattering cross sections of fast electrons on crystals-Yoshioka-equations-based fo...

Image and Spectrum Image Denoising under the local low Rank Assumption - Volume 24 Supplement - Jakob Spiegelberg, Juan Carlos Idrobo, Jan Rusz

Electron magnetic chiral dichroism (EMCD) is a promising technique to investigate local magnetic structures in the electron microscope. However, recognition of the EMCD signal, or also finding optimal parameter settings for given materials and sample orientations typically requires extensive simulations to aid the experiment. Here, we discuss how m...

Atomic resolution imaging and spectroscopy suffers from inherently low signal to noise ratios often prohibiting the interpretation of single pixels or spectra. We introduce local low rank (LLR) denoising as tool for efficient noise removal in scanning transmission electron microscopy (STEM) images and electron energy-loss (EEL) spectrum images. LLR...

The complex interplay of elastic and inelastic scattering amenable to different levels of approximation constitutes the major challenge for the computation and hence interpretation of TEM-based spectroscopical methods. The two major approaches to calculate inelastic scattering cross sections of fast electrons on crystals—Yoshioka-equations-based fo...

This paper demonstrates how Signal Subspace Sampling (SSS) is an effective pre-processing step for Non-negative Matrix Factorization (NMF) or Vertex Component Analysis (VCA). The approach allows to uniquely extract non-negative source signals which are orthogonal in at least one observation channel, respectively. It is thus well suited for processi...

Inelastic electron scattering is a consequence of mostly Coulomb interaction between electrons in the sample and electron beam and, as such, it is a nonlocal event. In atomic resolution experiments, it thus opens the following question: How far is the origin of the inelastic scattering signal that is observed when the electron beam is passing nearb...

Quantitative analysis of noisy electron spectrum images requires a robust estimation of the underlying background signal. We demonstrate how modern data compression methods can be used as a tool for achieving an analysis result less affected by statistical errors or to speed up the background estimation. In particular, we demonstrate how a multilin...

While the performance of magnetic tunnel junctions based on metal/oxide interfaces is determined by hybridization, charge transfer, and magnetic properties at the interface, there are currently only limited experimental techniques with sufficient spatial resolution to directly observe these effects simultaneously in real-space. In this letter, we d...

A selection of tensor decomposition techniques is presented for the detection of weak signals in electron energy loss spectroscopy (EELS) data. The focus of the analysis lies on the correct representation of the simulated spatial structure. An analysis scheme for EEL spectra combining two-dimensional and n-way decomposition methods is proposed. In...

Key to the advancement of magnetic materials is the exploitation of emergent magnetic behavior arising from nanoscale confinement. A comprehensive understanding of this behavior requires a technique capable of resolving it and mapping its relationship to parallel nanoscale property domains such as local structural imperfections, chemical environmen...

X‐ray magnetic circular dichroism is a well‐established method to study element specific magnetic properties of a material, while electron energy‐loss magnetic chiral dichroism (EMCD), which is the electron wave analogue to XMCD, is scarcely used today. Recently discovered electron vortex beams, which carry quantized orbital angular momenta (OAM) L...

it has been theoretically argued that atomic‐size electron probes with customized phase distributions can detect electron magnetic circular dichroism (EMCD) [1]. Based on this prediction we have recently shown that deliberately aberrated electron probes in scanning transmission electron microscopy (STEM) can be utilized to obtain chiral dichroic si...

Electron magnetic circular dichroism (EMCD) in the transmission electron microscope (TEM) is a relatively immature experimental technique, though it has already demonstrated quantitative results with a spatial resolution superior to what can be achieved with X‐rays [1]. The main obstacles for widely sharing the method for routine use are (i) low SN...

Despite various efforts, it remains an experimentally challenging task to access magnetic properties at (sub) nanoscale. One route towards a direct measurement of magnetism is the measurement of electron magnetic circular dichroism (EMCD) [1]. Being based on the measurement of electron energy loss (EEL) spectra, EMCD can in principle be measured at...

A set of geometric data decomposition methods is discussed. In particular, randomized vertex component analysis (RVCA), an extension of vertex component analysis (VCA) for the application to noisy data is established. Minimum volume simplex analysis (MVSA), a recent technique for the extraction of endmembers in the absence of pure pixels, is presen...

Nature Communications 7: Article number: 12672 (2016); Published: 31 August 2016; Updated: 5 October 2016 In the originally published version of the Article, an incorrect reference number was cited. The final sentence of the first paragraph of the Introduction section should read: ‘The likely reasons are technical difficulties in generating atomic...

Principal component analysis (PCA) is among the most commonly applied dimension reduction techniques suitable to denoise data. Focusing on its limitations to detect low variance signals in noisy data, we discuss how statistical and systematical errors occur in PCA reconstructed data as a function of the size of the data set, which extends the work...

Rapid development of magnetic nanotechnologies calls for experimental techniques capable of providing magnetic information with subnanometre spatial resolution. Available probes of magnetism either detect only surface properties, such as spin-polarized scanning tunnelling microscopy, magnetic force microscopy or spin-polarized low-energy electron m...

div class="title">Mapping Magnetic Ordering With Aberrated Electron Probes in STEM - Volume 22 Issue S3 - Juan Carlos Idrobo, Ján Rusz, Jakob Spiegelberg, Michael A. McGuire, Christopher T. Symons, Ranga Raju Vatsavai, Claudia Cantoni, Andrew R. Lupini

Although magnetism originates at the atomic scale, the existing spectroscopic techniques sensitive to magnetic signals only produce spectra with spatial resolution on a larger scale. However, recently, it has been theoretically argued that atomic size electron probes with customized phase distributions can detect magnetic circular dichroism. Here,...

Mapping Magnetic Properties of Materials At Atomic Spatial Resolution - Volume 21 Issue S3 - Jan Rusz, Juan-Carlos Idrobo, Alexander Edstrom, Jakob Spiegelberg, Somnath Bhowmick

In the framework of the slice transition operator technique, a general multislice theory for electron scattering in crystals is developed. To achieve this generalization, we combine the approaches for inelastic scattering derived by Yoshioka [J. Phys. Soc. Jpn. 12, 6 (1957)] and backscattering based on the formalism of Chen and Van Dyck [Ultramicro...

We examine the validity of the rotating wave approximation (RWA) in non-adiabatic holonomic single-qubit gates proposed in [New J. Phys. {\bf 14}, 103035 (2012)]. We demonstrate that RWA is widely valid but may lead to a sharp decline in fidelity for rapid gate implementation and small energy separation between the excited and computational states....