South Korea followed the textbook principles of epidemiology. Kept deaths 40 times lower all the way till 75% of population fully vaccinated. This is success.

A kinesin is a protein supporting several cellular functions including mitosis, meiosis and transport of cellular cargo bit.ly/2jEGA6P [source of the gif: buff.ly/2vGJDBN]

Learn about the significance of overparametrization in neural networks, the universal law of robustness, and what “A Universal Law of Robustness via Isoperimetry” means for future research in this short video with Sebastien Bubeck: youtu.be/ujMvnQpP528.

PartImageNet: A Large, High-Quality Dataset of Parts abs: arxiv.org/abs/2112.00933 a large, high-quality dataset with part segmentation annotations. It consists of 158 classes from ImageNet with approximately 240000 images

OMFG My 10 year old just asked Alexa on our Echo for a challenge and this is what she said.

The call for papers for the ICLR 2026 Workshop on the Elements of Reasoning: Objects, Structure, and Causality (OSC) is out! #osc2022 Submit your work by 25 Feb (end of day, AoE) 👉 objects-structure-causality.github.io

Stop funding research (the way you do) Because it is wasteful, unethical and unscientific A thread (1/16)

I’ve been using #AnnotatedEquations in my recent papers. I think it really adds to the readability and understanding of the math. Here are some examples. It uses #tikz in #latex. Let me know if you like it. Happy for any feedback. [GitHub link: next tweet] #AcademicChatter +

The TMLR team is happy to announce that we are now accepting submissions! jmlr.org/tmlr/news/2022…

15 bookmark-worthy tools for researchers that cost you nothing and will save you time (& nerves) when researching, writing and publishing. A thread. 🧵 #AcademicTwitter #AcademicChatter

Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks Miquel Martí i Rabadán,@alepieropan, Hossein Azizpour, Atsuto Maki tl;dr: use weak/strong aug consistency for segmentation. #kornia used for augmentation arxiv.org/abs/2210.09919

#SeRC is organizing a #Workshop on November 30th to discuss the results of our projects on #ML and #eScience, including #FluidMechanics, #Climate, #LifeScience, etc. All the information (also with #online attendance): azizpour.github.io/sercw22/ Hossein Azizpour, Hedvig Kjellström

TMLR (Transactions on Machine Learning Research) just put out a report on their first (almost) year of operations. Here's a 🧵 with some of the highlights. docs.google.com/document/d/1tJ… 1/n

📰BREAKING!!! We have used #DeepReinforcementLearning (#DRL) for #Turbulence #Control in #Channel flow, and we have obtained #higher #DragReduction (46%) than the classical #OppositionControl (20%)!! #preprint: lnkd.in/db62jfpJ #Code: To be released soon!

Deep Double Descent via Smooth Interpolation Matteo Gamba, Erik Englesson, Mårten Björkman, Hossein Azizpour. Action editor: Sebastian Stich. openreview.net/forum?id=fempQ… #overparameterized #deep #interpolated

🚀 Daily, we marvel at #AI progress, yet one human ability stands unchallenged: immediate adaptation. 💡 Today, we're closing that gap! Introducing 💀 To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation. An AI that shifts gears as swiftly as we do! #ICCV2023

A nice way to end the year with some data: 66 Good News Stories You Didn't Hear About in 2023 futurecrunch.com/goodnews2023/