RANDALL BALESTRIERO

I have been doing research in learnable signal processing since 2013, in particular with learnable parametrized wavelets [ref] which has then been extended for deep wavelet transforms [ref]. The latter has found many applications e.g. in the NASA's Mars rover [ref] for marsquake detection. In 2016 when joining Rice University for a PhD with Prof. Richard Baraniuk, I broadened my scope to explore Deep Networks from a theoretical persepective by employing affine spline operators [ref1, ref2]. This led me to revisit and improve state-of-the-art methods e.g. batch-normalization [ref] or generative networks [ref1, ref2]. In 2022 when joining Meta AI Research (FAIR) for a postdoc with Prof. Yann LeCun, I further enlarged my research interests e.g. to include self-supervised learning or biases emerging from data-augmentation and regularization leading to many publications... Fighting to bridge the gaps and cracks between Deep Learning theory and best practices by developping realistic..