Recent publications

With PCX, we introduce a method that summarizes similar single explanations via prototypical ones. As such, we can understand the whole model behavior quickly and in detail. PCX further allows to validate individual predictions by communicating...

Our new method, AttnLRP, is the first to faithfully and holistically attribute not only input but also latent representations of transformer models with the computational efficiency similar to a single backward pass. We demonstrate that our...

While recent studies show high accuracy in the classification of Alzheimer's disease using deep neural networks, the underlying learned concepts have not been investigated. We separated Alzheimer's patients (n=117) from normal controls (n=219) by...

Neurons in deep neural networks can act polysemantically, meaning that they encode for multiple (unrelated) features. As such, understanding the inner workings of machine learning models becomes more difficult. We present PURE to turn...

The paper proposes a class of distributed algorithms where the consensus step is implemented in a scalable and truly decentralized fashion using a novel communication protocol based on the “over-the-air” function computation (OTA-C)...

Heatmaps generated on inputs of image classification networks via explainable AI methods have been observed to resemble segmentations of input images in many cases. We apply the "Right for the Right Reason" paradigm of imposing additional losses...

Centralized or distributed precoding? This is perhaps the most heated debate in the Cell-free Massive MIMO literature. However, in this work we argue that the best option may actually be a mix of the two. The reason is that centralized precoding,...

We present a hybrid integrated photonic circuit with tunable arrayed waveguide gratings (AWGs) as (DE-)MUX stages for optical modulators for use in parallel convolution processing. The simulated and fabricated AWGs have 16 outputs with 500 GHz...

For highly loss-sensitive applications of photonic integrated circuits (PICs), efficient coupling of high-confinement silicon nitride (Si3N4) passive waveguides with active optoelectronic components and standard single mode fibers (SMF) is...

We present an innovative integrated photonic sensor for point-of-care applications using silicon nitride (Si3N4) tailored for biofluids analysis in the 850nm optical band. With microring resonators (MRR) we detect smallest changes in the...

We present a novel approach that combines the high rendering quality of NeRF-based 3D-aware GANs with the flexibility and computational advantages of 3DGS. By training a decoder that maps implicit NeRF representations to explicit 3D Gaussian...

For time series data, where the input itself is often not interpretable, dedicated XAI research is scarce. In this work, we put forward a virtual inspection layer for transforming the time series to an interpretable representation and allows to...

This paper presents an optimization framework for analog beamforming in monostatic integrated sensing and communication (ISAC), which maximizes sensing detection under self-interference cancellation via superiorized projections onto convex...

Exploring near-field propagation is vital for 6G technologies like intelligent reflecting surfaces (IRS). Unlike far-field models, near-field models offer accuracy critical for applications such as beamforming and multiple-access, enhancing...

We propose and demonstrate a novel O-band wavelength division multiplexing (WDM) optical transceiver enabled by the hybrid photonic integration of Indium Phosphide (InP) components into a Polymer based photonic motherboard c! alled PolyBoard. The...

Campus networks have become a major market segment for cellular communication technology, providing a flexible communication infrastructure to meet the specific de-pendability and performance requirements of industry verticals. With the emerging...

This paper introduces a new concept and outlines the implementation of an AR application designed for mobile devices. It can visualize real-time environmental data from various Open Data platforms. The scientific contribution lies in diverse...

We present a fiber-pigtailed hybrid photonic circuit comprising nonlinear waveguides for photon-pair generation and a polymer interposer reaching 68 dB of pump suppression and photon separation based on a polarizing beam splitter with > 25 dB...

We present iV2V and iV2i+, two machine-learning datasets for industrial wireless communication. The datasets cover sidelink and cellular communication involving autonomous robots together with localization and sensing data, which can be used to...

Sub-THz communication, vital for 6G, requires densification and multi-connectivity to overcome signal blockage. Efficient beam and user scheduling are crucial. We propose a low-complexity scheduling scheme using machine learning (ML), employing a...

This paper addresses the massive MIMO unsourced random access problem in quasi-static Rayleigh fading. Our proposed coding scheme combines an outer LDPC code with an inner SPARC code, integrating channel estimation, single-user decoding, and...

We propose a novel representation of virtual humans for highly realistic real-time animation and rendering in 3D applications. We learn pose dependent appearance and geometry from highly accurate dynamic mesh sequences obtained from...

We utilized an EEG dataset of steady-state visual evoked potentials in which participants were presented with human face images of different stylization levels. Assessing neuronal responses, we found a non-linear relationship between SSVEP...

AI has proven effective in classifying skin cancers using dermoscopy images. However, clinical application is limited when algorithms are not well-trained, or there is a lack of clinical context (e.g., medical history). The increasing use of...

This article advances the theoretical foundations of user-centric cell-free massive MIMO networks. In particular, by means of a novel UL-DL duality principle for fading channels, it settles the optimality of the recently developed “team MMSE”...

Traditional signal processing methods often fall short compared to data-driven approaches in a signal separation problem that involves co-channel signals, where the energy content of the interference component overlaps with the transmitted signal...

This paper compares the sub-THz radio channel characteristics at 158 GHz and 300 GHz in a shopping mall scenario by extracting three different path loss models and various channel parameters.

We present a novel method ensuring the right reasons on the concept level by reducing the model's sensitivity towards biases through the gradient. When modeling biases via Concept Activation Vectors, we highlight the importance of choosing robust...

Paper studies a class of distributed fixed-point algorithms over truly decentralised and unbalanced graphs, supporting a broad class of communication systems, including a novel OTA computation protocol that implements consensus without any...

Future networks aim for higher data rates through cell densification, leading to frequent handovers and increased signaling overhead for moving user equipment (UE). This paper presents an optimization scheme using predicted channel state...