Very clever by @rastrau to be displaying elevation data for Europe in this tile format. I hadn't seen this done before. Source: https://t.co/z3NujuQqs3 pic.twitter.com/5v9LvLXQF0

— Simon Kuestenmacher (@simongerman600) March 19, 2021

Actually, I'd argue it's impossible.

So let's not pretend that AI can do it. https://t.co/kFu4HutH2U

— Cathy O'Neil (@mathbabedotorg) March 19, 2021

Learning to Resize Images for Computer Vision Tasks
pdf: https://t.co/TlwiWtXz5r
abs: https://t.co/cQyXmFJeVV pic.twitter.com/6YRg1FoDLr

— AK (@ak92501) March 19, 2021

All NLP Tasks Are Generation Tasks: A General Pretraining Framework
pdf: https://t.co/Zhdx33AUgL
abs: https://t.co/ibJdjVbIsD pic.twitter.com/8OPdrHVCPn

— AK (@ak92501) March 19, 2021

When you rewrite stuff in Rust, you add a --load_fast=True flag, apparently https://t.co/l4L7XgyHtt

— François Chollet (@fchollet) March 19, 2021

⚠️ New in the v0.0.7 release of `huggingface_hub` ⚠️

Community member @7vasudevgupta added the ability to mix-in a class named ModelHubMixin to *any PyTorch model* to be able to save, upload it, and load it from the https://t.co/ZygXVJ8qCM hub:

🔥🔥https://t.co/1jZ4Y6GPwo pic.twitter.com/mfCwH0f5sF

— Julien Chaumond (@julien_c) March 18, 2021

ALADIN: All Layer Adaptive Instance Normalization for Fine-grained Style Similarity
pdf: https://t.co/K7homQW2GM
abs: https://t.co/jTp3DS4u7o pic.twitter.com/NUBfHbZZWE

— AK (@ak92501) March 18, 2021

Happy to share our tutorial's video on "Causal Fairness Analysis" in the @FAccTConference this year (joint work w/ Drago Plecko & @junzhez): https://t.co/Nn67gjczLw, and the slides: https://t.co/nyd0bmNJse. Special thanks to the organizers, @tiberiocaetano & @zacharylipton .)

— Elias Bareinboim (@eliasbareinboim) March 17, 2021

Freshly merged in the @scikit_learn main branch: periodic spline features:https://t.co/g6MTtmMKSY

Avoids introducing a jump between Dec 31st and Jan 1st when doing non linear, smooth feature engineering on a "day of year" input feature for instance.

Thanks Malte Londschien! pic.twitter.com/g638Sl6sZa

— Olivier Grisel (@ogrisel) March 17, 2021

Is it Enough to Optimize CNN Architectures on ImageNet?
pdf: https://t.co/zC5jToLTto
abs: https://t.co/oIYWstLrIf pic.twitter.com/RrEHLNJEqa

— AK (@ak92501) March 17, 2021

Multi-view subword regularization is simple but yields consistent improvements over pre-trained multilingual models. The best thing: It only needs to be applied during fine-tuning.

Paper: https://t.co/gxTgbzVvWN
Code: https://t.co/FqUyZgEnOQ https://t.co/sTFxot6yan

— Sebastian Ruder (@seb_ruder) March 16, 2021

Few-Shot Text Classification with Triplet Networks, Data Augmentation, and Curriculum Learning
pdf: https://t.co/uyzxL1tu4L
abs: https://t.co/qX90yC7nAH pic.twitter.com/Y15zxoaJo8

— AK (@ak92501) March 16, 2021