I'm Ferenc Huszár. a machine learning researcher at Twitter. I did my PhD in Bayesian ML at Cambridge University with Carl Rasmussen, Máté Lengyel and Zoubin Ghahramani. I'm interested in probabilistic inference, generative models, representation learning, information theory, deep learning and many more things.
Follow me on twitter @fhuszar or here's my Google scholar profile if that's more what you're after.
I started this blog some time ago but now it kind of has its own life. inFERENCe got started when, in 2015, I returned to machine learning research after a 3-year stint as a data scientist. I basically slept through the deep learning revolution. In those three years, many things happened, so I had to play catch up.
Initially, these blog posts helped me understand the body of literature I have missed: generative adversarial networks, variational autoencoders, representation learning, etc. Nowadays, I continue reading and writing about current papers, trying to reinterpret them and find connections to things I know, usually ending up with some kind of KL divergence.
If you're new here and want to get a taste, you may want to start with these crowd favourites:
- dilated convolutions and Kronecker products
- Gaussian Distributions as Soap Bubbles
- Everything that works works because it's Bayesian
- GANs are broken in more than one way
Twitter, Magic Pony
I work at Twitter as a member of the London-based Magic Pony team. Over the past few years we have worked on low-level vision, video compression, superresolution, image and video classification and other machine learning problems related to media on Twitter.
We ended up at Twitter via the acquisition of our startup Magic Pony Technology. Prior to Magic Pony I worked as part of the investment team at Balderton Capital, an early-stage venture capital firm in London (hence the .vc domain name which I'm too lazy to change).