I’m a research scientist at Google DeepMind in London, UK, where I work chiefly on generative modelling at scale. Before that, I was a PhD student at Ghent University in Belgium. On this blog, I mostly write about machine learning, deep learning and generative models.
My PhD research topic was learning hierarchical representations of musical audio signals for classification and recommendation, with a focus on deep learning and feature learning. More information can be found on the Research page. Since then I have worked on audio generation (e.g. WaveNet), AlphaGo, representation learning, and generative modelling with autoregressive models, GANs and diffusion models, including Lyria, Imagen 2, Imagen 3 and Veo.
During my PhD, I was also quite active on Kaggle. A few write-ups about my experience with Kaggle competitions are available on this blog. I also wrote a paper about my winning solution for an astronomy competition.
I was one of the lead developers of the Theano-based neural network library Lasagne.