Franciskus Xaverius Erick

PhD student. Representation learning and uncertainty quantification @UniFAU @BorderlessSci @BiomedIAICL

[Gabriel Peyré @gabrielpeyre]

Stein's unbiased risk estimate (SURE) is an almost magical formula that enables the computation of the mean squared error of a denoiser (used, for example, in denoising score matching) using only the noisy observation y, without requiring the clean data x. https://t.co/NwmXkmRC15

[shaurya @shauseth]

me and my supervisor discussing how our predictions are worse than random

[Frank Nielsen @FrnkNlsn]

🎓Kullback-Leibler divergence between densities of an exponential family = reverse Bregman divergence wrt the cumulant function

🎉Kullback-Leibler divergence between non-normalized densities = reverse Bregman divergence wrt the partition function

👉 https://t.co/bmN5jDvwza

[Sander Dieleman @sedielem]

New blog post! Some thoughts about diffusion distillation. Actually, quite a lot of thoughts 🤭 Please share your thoughts as well!

https://t.co/JZyRsjC25v

[KaiAI.AI @kaikim29]

Shoutout to the legend who answers literally every single PyTorch question out there!

[Pablo Samuel Castro @pcastr]

MICo: Improved representations via sampling-based
state similarity for MDPs

Our #NeurIPS2021 paper introduces a new loss that improves your RL agents!

📜Paper: https://t.co/MXN0t4ytha
💻Blog: https://t.co/pw3B1D8XpF
🐍Code: https://t.co/XbLefgQAhh

1/🧵

[Jia-Bin Huang @jbhuang0604]

How to do research with my mentors effectively?

I get this question frequently in my open office hours. I am still learning as well but I hope sharing my ✌💰 may be helpful to some.

Key idea ➡️ **Help them help you!**

How? Check out the thread 🧵

[AutoML.org @AutoML_org]

Unsure which arch to use for your deep ensemble? Why settle for one? Neural Ensemble Search constructs ensembles with varying network archs

Paper: https://t.co/EImmbXUKD9
Code: https://t.co/1uhTwm0oWG

Work by @ShehZaidi @ZelaArber Thomas Elsken @cholmesuk @FrankRHutter @yeewhye

[Bernhard Egger @VisionBernie]

Sorry if somebody did this one before - but the field is growing so fast, there is no way I can keep track of it!
While making this I constantly felt like somebody is scooping me - or was I already scooped?
#ComputerVision
#CVPR2021
#TypesofPaper

[Jia-Bin Huang @jbhuang0604]

Types of Computer Vision Paper (source: @xkcd)

[Joan Puigcerver @joapuipe]

Check out what we’ve been working on for the last months: We decouple the model size and the compute cost in a Vision Transformer backbone by using Sparse MoE layers. These have been popularised in NLP, and they are fantastic for Vision too! https://t.co/AdDwJKvkBk

[Neil Houlsby @neilhoulsby]

[3/3] Towards big vision

While dense models are still the norm, sparse MoE layers can work well too!

Large Vision-MoEs (15B params) can be trained to high performance relatively efficiently, and can even prioritize amongst patches (see duck).

https://t.co/0WHyJTlGG5

...