Durk Kingma (@dpkingma) 's Twitter Profile
Durk Kingma

@dpkingma

@AnthropicAI. Prev. @Google Brain/DeepMind, founding team @OpenAI. Computer scientist; inventor of the VAE, Adam optimizer, and other methods. ML PhD.

ID: 25263396

linkhttps://dpkingma.com calendar_today19-03-2009 09:32:31

676 Tweet

47,47K Followers

401 Following

Yisong Yue (@yisongyue) 's Twitter Profile Photo

Congratulations to Durk Kingma and Max Welling for winning the inaugural ICLR Test of Time Award for their amazing work on Auto-Encoding Variational Bayes, the paper that proposed Variational Autoencoders! arxiv.org/abs/1312.6114

Durk Kingma (@dpkingma) 's Twitter Profile Photo

Thanks to the ICLR Award Committee! And thank you for the kind words, Max! You were the perfect Ph.D. advisor and collaborator, kind and inspiring. I really couldn't have wished for better.

Sirui Xie (@siruixie) 's Twitter Profile Photo

📢 Excited to share EM Distillation (EMD), a maximum likelihood method that distills pretrained diffusion models to one-step generators. EMD gracefully interpolates between mode-seeking and mode-covering KL to better capture the teacher's distribution. arxiv.org/abs/2405.16852

📢 Excited to share EM Distillation (EMD), a maximum likelihood method that distills pretrained diffusion models to one-step generators. EMD gracefully interpolates between mode-seeking and mode-covering KL to better capture the teacher's distribution.
 
arxiv.org/abs/2405.16852
Durk Kingma (@dpkingma) 's Twitter Profile Photo

The recording of our talk for the ICLR'24 test-of-time award (with Max Welling) is now available online: iclr.cc/virtual/2024/t… Biggest live audience I've ever spoken to, with >2000 attendees 😅. But it was a lot of fun!

Sander Dieleman (@sedielem) 's Twitter Profile Photo

In arxiv.org/abs/2303.00848, Durk Kingma and Ruiqi Gao had suggested that noise augmentation could be used to make other likelihood-based models optimise perceptually weighted losses, like diffusion models do. So cool to see this working well in practice!

Durk Kingma (@dpkingma) 's Twitter Profile Photo

Great blogpost by Ruiqi (and other GDM ex-colleagues), clearly explaining the the connection between flow matching and diffusion models. Super happy they took the time to explain this topic, there's confusion on this topic, I think many will find this quite valuable!

Durk Kingma (@dpkingma) 's Twitter Profile Photo

👇 Great work led by Yushun (Yushun Zhang) introducing Adam-mini, a version of Adam that, surprisingly, reduces Adam's memory requirement by 50% (!), without negatively affecting convergence rates. Please read Yushun's thread for details!