Are you (or is someone you know) teaching AI / ML / Deep Learning this year? My forthcoming book (freely available at https://t.co/hqRA1xUPkk) will save you a lot of time. This thread will show you why.
SAM, the groundbreaking segmentation model from @Meta is now in available in 🤗 Transformers!
What does this mean?
1. One line of code to load it, one line to run it
2. Efficient batching support to generate multiple masks
3. pipeline support for easier usage
More details: 🧵
@airindiain Yes, after 10 hours of waiting at the airport, with no sleep. The flight should have reached its destination by now.
Thanks for this useless information.
New update to deep learning textbook:
https://t.co/43UgPR6C4X
Many small improvements plus a new chapter on variational autoencoders. This also provides most of the theory for training diffusion models (diffusion chapter coming next week). Feedback please!
I once killed a $125 million deal by being “too honest.” There are many ways to lose deals, lest you think last week’s story was my only painful blunder. This one hinged on a fateful encounter with Jerry Yang and involved Paul Graham (@paulg). Here’s what happened.... (1/n)
What is in common between CNNs, GNNs, LSTMs, Transformers, DeepSets, mesh CNNs? In a new post with @joanbruna@TacoCohen@PetarV_93 we show this zoo of neural nets can be seen through the lens of symmetry. #geometricdeeplearning is all you need!
https://t.co/BMYfaz7sBK
We run one of the largest Kubernetes clusters in the world. Having one massive cluster lets us iterate rapidly at small scale, and also scale up large models without changing code.
Tremendous effort by our supercomputing team (who is hiring: https://t.co/El9s3HjrpK!).
Possible scenario: It's 2025. The epidemic is over, but most big tech cos are still significantly remote, because powerful employees moved to Jackson Hole and don't want to move back. A startup emerges that's intensely not-remote, and beats one of the incumbents by moving faster.
We are releasing Detection Transformers (DETR), an important new approach to object detection and panoptic segmentation. It’s the first object detection framework to successfully integrate Transformers as a central building block in the detection pipeline.
https://t.co/mvDSMstiRl