I'm a PhD student in physics and machine learning currently interested in understanding why deep learning works by bridging the gap between theory and practice. I identify myself as a machine learning experimentalist. My work focuses on the rigorous study of existing deep neural networks to discover fundamental recurring laws that deepen our understanding of why deep learning works and can help and guide, on the practitioners' side, the design of novel models and architectures and, on the theoretical side, to ask the correct questions.