Advances in Classifier-Free Guidance
Classifier-Free Guidance (CFG) is essential for conditional diffusion models. This post covers CFG basics and two NeurIPS 2025 papers with new theoretical insights and improvements.
Blog Posts
ICLR 2020 impressions and paper highlights
Having just "visited" my first virtual conference, ICLR 2020, I wanted to talk about my general impression and highlight some papers that stuck out to me from a variety of subfields.
Spotify Internship Report
From June to September 2019, I took a break from my ongoing PhD and worked as a Research Intern at Spotify in London.
Bounded output regression with neural networks
When neural networks need bounded outputs (like audio in [-1,1]), tanh or clipping during training causes vanishing gradients. Use linear outputs during training and clip only at test time.
ISMIR 2018 - Paper Overviews
This year's ISMIR was great as ever, this time featuring lots of deep learning, new datasets, and a fantastic boat tour through Paris!