Revisiting ResNets: Improved Training and Scaling Strategies
pdf: https://t.co/Pn5cU2SVkB
abs: https://t.co/icpnuFwmXU pic.twitter.com/bA0E1GWR5z

— AK (@ak92501) March 16, 2021

Facebook AI has built TimeSformer, a new architecture for video understanding. It’s the first based exclusively on the self-attention mechanism used in Transformers. It outperforms the state of the art while being more efficient than 3D ConvNets for video.https://t.co/8mQ2rMgcDo pic.twitter.com/dBpbT3UJRx

— Facebook AI (@facebookai) March 15, 2021

Very sobering chart by @gzeromedia shows how a decade of war has crushed Syria. Source: https://t.co/2kBoomECFj pic.twitter.com/oqd7NPHTS6

— Simon Kuestenmacher (@simongerman600) March 14, 2021

ConSelfSTransDRLIB: Contrastive Self-supervised Transformers for Disentangled Representation Learning with Inductive Biases is All you need, and where to find them.

The current state of deep learning research summarized in one sentence.

(Credit: https://t.co/RTuht7Lkj0)

— Sebastian Raschka (@rasbt) March 13, 2021

"The firings have also been an unmitigated PR disaster for the tech giant. From a distance, the story sounds like “Google fires the leadership of its AI ethics team for doing their jobs.” "https://t.co/j7Hb3NZyJk

— Emily M. Bender (@emilymbender) March 12, 2021

🔥Fine-Tuning @facebookai's Wav2Vec2 for Speech Recognition is now possible in Transformers🔥

Not only for English but for 53 Languages🤯

Check out the tutorials:
👉 Train Wav2Vec2 on TIMIT https://t.co/33Bx8Nj4mN
👉 Train XLSR-Wav2Vec2 on Common Voicehttps://t.co/xOoEQV3Krn pic.twitter.com/rxp2hAbaLS

— Hugging Face (@huggingface) March 12, 2021

CUAD: A dataset with over 13,000 annotations for hundreds of legal contracts that have been manually labelled by legal experts, to serve as a benchmark for contract understanding.

Discussion: https://t.co/qP4dC40Z8l
GitHub: https://t.co/Kj7NZs3qW0
Paper: https://t.co/RTbXxbs06a pic.twitter.com/IZthZPAPs0

— hardmaru (@hardmaru) March 12, 2021

Good news! My lecture on "What is Causal Inference" is already available on youtube, free for all: https://t.co/RFRxUXnAAd
Enjoy!
And, if there are questions, do not hesitate, preferably after reading #Bookofwhy, for the obvious reason that the answer is very likely to be there.

— Judea Pearl (@yudapearl) March 11, 2021

I'm not too old to remember when, at a panel discussion where Yann LeCun was asked about ethics and fairness at Facebook, he responded cheerfully, "that's above my paygrade!".

Now he's tweeting about fairness at Facebook.

I wonder what changed. Cynically I think it's pure PR. https://t.co/P6PhjQF8lJ

— Cathy O'Neil (@mathbabedotorg) March 11, 2021

It's been a rough couple news days if you care about the social impact of tech.😞

I've pulled together a couple articles below, but to summarize: AI applications are genuinely hurting people, companies are profiting off it & we have increasing evidence self regulation is a myth.

— Rachael Tatman (@rctatman) March 11, 2021

A blog post on algorithmic fairness work at Facebook,
and a research paper on the topic.

Paper: "Fairness On The Ground" https://t.co/InVLlshw7Q

Blog post: "What AI fairness in practice looks like at Facebook"https://t.co/sUmXKyAYu9

— Yann LeCun (@ylecun) March 11, 2021

Pretrained Transformers as Universal Computation Engines https://t.co/dsmqvAEDCB

— /MachineLearning (@slashML) March 11, 2021