Last night I had a vivid dream that I received an email with a checkout link to purchase the Banksquiat print.

One of the peculiarities of life is celebrating the birth of your newborn and coming to terms with a family french bulldog passing shortly thereafter.

We just released PyTorch3D, a new toolkit for researchers and engineers that’s fast and modular for 3D deep learning research: ai.facebook.com/blog/-introduc…

Taylor swift as different types of imaging studies. A thread.

How to Reduce Infections Without Closing Businesses nytimes.com/interactive/20… Excited to see our collaboration with NYT out!

Self-supervised learning: The dark matter of intelligence ai.facebook.com/blog/self-supe…

when you definitely understand NFTs

QUIZ TIME! What tensor memory format works faster for vision models in PyTorch? A. Makes no difference B. Whatever the default is C. Channels-Last (NHWC) D. Contiguous (NCHW) The answer is in the next tweet 👇🏽 1/6

Amazing mind-bending conversation with AGI.Eth on a vision to change the world with putting knowledge on the blockchain to educate all with a DAO and monetization 🤩 Host: codechick.eth Questions by: Shea Lewis cyrus #ArtificialIntelligence #NFT x.com/i/spaces/1OwxW…

#Breaking: Today we announced positive Phase III results for our subcutaneous #immunotherapy in previously treated metastatic non-small cell lung cancer. Learn more: gene.com/media/press-re…

Join mike0sv next week for his workshop #TWIMLcon on Codifying, Standardizing, and Serving Production-Ready ML Models with MLEM. Tickets available here: twimlai.com/conf/twimlcon/…

Have Yourself A Merry Little Christmas ♥️

Giving Away a TIER 3 SEWER PASS 💎 To someone who retweets this tweet! Can’t wait for the game Bored Ape Yacht Club 🍌 I’ll pick a random winner at 5pm pt tmrw !!!

Towards Universal Cell Embeddings. My talk at the Broad Institute about our recent advancements in ML for biomedicine: i.stanford.edu/~jure/pub/talk… Thank you Dr. Caroline Uhler for hosting!

Everything you ever wanted to know about Self-Supervised Learning but were afraid to ask. A giant cookbook of SSL recipes. By a large crowd from Meta-FAIR with various academic collaborators led by Randall Balestriero and Mark Ibrahim. arxiv.org/abs/2304.12210

Someone asked Midjourney to create female warriors from each country 🔥 1. Greece

This year, I trained 48 models for 173 hours on Weights & Biases using my 3090s with a custom water loop in a Fractal mid-tower. My #longestrun, mask2former_r50_1xb6-ssl_split1, took 16 hours.

Excited to share that RelBench: Relational Deep Learning Benchmark paper was accepted to NeurIPS. Summarizing the study in one picture 📷 relbench.stanford.edu

Linoleic acid is the most abundant polyunsaturated fatty acid in the Western-style diet and has been implicated in promoting inflammation and cancer. In Science, researchers report a mechanism by which linoleic acid can activate the mTORC1 protein kinase, which coordinates

New paper from FAIR+NYU: Q: Is language supervision required to learn effective visual representations for multimodal tasks? A: No. ⬇️⬇️⬇️