arXiv subjects

Björn Kampa

Publications and source records attributed to Björn Kampa.

3 recordsLinked to original sources

arXiv2026

Rewiring of adult neural networks - learning from structural plasticity across multiple scales

Synaptic connections are continuously formed and eliminated even in adult neuronal networks. This process - called structural plasticity - allows networks to change their topology to learn and adapt to their environment or compensate for lesions. Yet, in contrast to synaptic plasticity, structural plasticity is less well studied and understood. We here present the current state of knowledge as well as open problems in the field along three guiding questions: (1) What regulates structural plasticity in biological neural networks? (2) What are the functional consequences and advantages of structural plasticity in adult neural networks? (3) How can our understanding of structural plasticity be translated into clinical and technological applications? We argue that understanding structural plasticity is a problem involving multiple temporal and spatial scales. Solving this problem thus requires collaborative efforts of experimentalists, theoreticians as well as computational modelers, and has the potential for comprehensive neuroscientific insights that can guide the development of medical applications, artificial neural networks as well as neuromorphic computing.

q-bio.NC↗

arXiv2025

Towards Robust and Generalizable Gerchberg Saxton based Physics Inspired Neural Networks for Computer Generated Holography: A Sensitivity Analysis Framework

Computer-generated holography (CGH) enables applications in holographic augmented reality (AR), 3D displays, systems neuroscience, and optical trapping. The fundamental challenge in CGH is solving the inverse problem of phase retrieval from intensity measurements. Physics-inspired neural networks (PINNs), especially Gerchberg-Saxton-based PINNs (GS-PINNs), have advanced phase retrieval capabilities. However, their performance strongly depends on forward models (FMs) and their hyperparameters (FMHs), limiting generalization, complicating benchmarking, and hindering hardware optimization. We present a systematic sensitivity analysis framework based on Saltelli's extension of Sobol's method to quantify FMH impacts on GS-PINN performance. Our analysis demonstrates that SLM pixel-resolution is the primary factor affecting neural network sensitivity, followed by pixel-pitch, propagation distance, and wavelength. Free space propagation forward models demonstrate superior neural network performance compared to Fourier holography, providing enhanced parameterization and generalization. We introduce a composite evaluation metric combining performance consistency, generalization capability, and hyperparameter perturbation resilience, establishing a unified benchmarking standard across CGH configurations. Our research connects physics-inspired deep learning theory with practical CGH implementations through concrete guidelines for forward model selection, neural network architecture, and performance evaluation. Our contributions advance the development of robust, interpretable, and generalizable neural networks for diverse holographic applications, supporting evidence-based decisions in CGH research and implementation.

cs.CV↗

arXiv2024

A Wasserstein Graph Distance Based on Distributions of Probabilistic Node Embeddings

Distance measures between graphs are important primitives for a variety of learning tasks. In this work, we describe an unsupervised, optimal transport based approach to define a distance between graphs. Our idea is to derive representations of graphs as Gaussian mixture models, fitted to distributions of sampled node embeddings over the same space. The Wasserstein distance between these Gaussian mixture distributions then yields an interpretable and easily computable distance measure, which can further be tailored for the comparison at hand by choosing appropriate embeddings. We propose two embeddings for this framework and show that under certain assumptions about the shape of the resulting Gaussian mixture components, further computational improvements of this Wasserstein distance can be achieved. An empirical validation of our findings on synthetic data and real-world Functional Brain Connectivity networks shows promising performance compared to existing embedding methods.

cs.CE↗