In the last years, neural networks (NN) have evolved from laboratory environments to the
state-of-the-art for many real-world problems. It was shown that NN models (ie, their weights …

Neural Representations have recently been shown to effectively reconstruct a wide range of
signals from 3D meshes and shapes to images and videos. We show that, when adapted …

Learning representations of well-trained neural network models holds the promise to
provide an understanding of the inner workings of those models. However, previous work …

Parameter-Efficient Fine-Tuning is widely used to transfer models between different
domains. However, for some high-reliability-equipment, the degradation is at a slow rate and …

With growing size of Neural Networks (NNs), model sparsification to reduce the
computational cost and memory demand for model inference has become of vital interest for …

Parameter generation has struggled to scale up for a long time, significantly limiting its range
of applications. In this study, we introduce\textbf {R} ecurrent diffusion for large-scale\textbf …

This thesis addresses the challenge of understanding Neural Networks through the lens of
their most fundamental component: the weights, which encapsulate the learned information …

The availability of large-scale labeled datasets in remote sensing and Earth observation
accelerated the use of deep neural networks in this domain. In the standard workflow, data is …

Mechanistic interpretability is a field that aims to explain the behaviour of trained neural
networks by studying their learnt parameters. As most of this line of work is laborious and …

Self-supervised learning allowed us to train large task agnostic backbones, which can be
successfully finetuned for specialized downstream tasks with only little supervision. This …