Distributed Convex Optimization “Over-the-Air” in Dynamic Environments
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)...
Neuromorphic Wireless Device-Edge Co-Inference via the Directed Information Bottleneck
We consider neuromorphic wireless device-edge co-inference for edge intelligence applications. The directed information bottleneck principle is applied to extract the most relevant information for the task. The model demonstrates superior...
Understanding the (Extra-)Ordinary: Validating Deep Model Decisions with Prototypical Concept-based Explanations
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...
AttnLRP: Attention-Aware Layer-wise Relevance Propagation for Transformers
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...
Reactive Model Correction: Mitigating Harm to Task-Relevant Features via Conditional Bias Suppression
DNNs are prone to relying on spurious correlations in data, posing risks in critical applications. Post-hoc methods exist to mitigate this without retraining but can globally shift latent features distributions, harming model performance. We...
Model guidance via explanations turns image classifiers into segmentation models
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...
PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits
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...
Explainable Concept Mappings of MRI: Revealing the Mechanisms Underlying Deep Learning-Based Brain Disease Classification
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...
Unlocking the Potential of Local CSI in Cell-Free Networks with Channel Aging and Fronthaul Delays
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,...
Towards Bridging the Gap between Near and Far-Field Characterizations of the Wireless Channel
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...