In this thesis, we study two closely related directions: robustness and generalization in modern deep learning. Deep learning models based on empirical risk minimization are known to be often non-robust to small, worst-case perturbations known as adversari ...
Machine learning has transformed many fields and has recently found applications in chemistry and materials science. The small datasets commonly found in chemistry sparked the development of sophisticated machine learning approaches that incorporate chemic ...
Over the past years, image generation and manipulation have achieved remarkable progress due to the rapid development of generative AI based on deep learning. Recent studies have devoted significant efforts to address the problem of face image manipulation ...
Neural decoding of the visual system is a subject of research interest, both to understand how the visual system works and to be able to use this knowledge in areas, such as computer vision or brain-computer interfaces. Spike-based decoding is often used, ...
Marmoset monkeys encode vital information in their calls and serve as a surrogate model for neuro-biologists to understand the evolutionary origins of human vocal communication. Traditionally analyzed with signal processing-based features, recent approache ...
Metapopulation models have been instrumental in quantifying the ecological impact of landscape structure on the survival of a focal species. However, extensions to multiple species with arbitrary dispersal networks often rely on phenomenological assumption ...
The choice of the shape parameter highly effects the behaviour of radial basis function (RBF) approximations, as it needs to be selected to balance between the ill-conditioning of the interpolation matrix and high accuracy. In this paper, we demonstrate ho ...
Modern machine learning models use an ever-increasing number of parameters to train (175 x 10(9) parameters for GPT-3) with large datasets to achieve better performance. Optical computing has been rediscovered as a potential solution for large-scale data p ...
Training accurate and robust machine learning models requires a large amount of data that is usually scattered across data silos. Sharing, transferring, and centralizing the data from silos, however, is difficult due to current privacy regulations (e.g., H ...
