I’m super excited to share our work on AdA: An Adaptive Agent capable of hypothesis-driven exploration which solves challenging unseen tasks with just a handful of experience, at a similar timescale to humans. sites.google.com/corp/view/adap… See the thread for more details 👇 [1/N]

Very proud to announce the 3rd edition of the @m2lschool which will be held in Thessaloniki - 28 Aug to 2 Sept 2023 and will take place @actgreece. APPLY AT: m2lschool.org/application Deadline: February, 28th (or 600 applications) Don't miss this opportunity! 🔥⏱

Only a few days left to apply - don't miss the chance to learn from incredible speakers and network with outstanding researchers from all over the world!

Incredible step forward in image segmentation and scene understanding. Truly awesome!

Wow.

Wow! This is amazing, can't wait to see all the cool applications that will come out of this!

Fourth time's a charm!! 🤩 M2L applications are open ✍️ Don't miss this unique opportunity to learn from the best speakers, network with incredibly talented researchers, and visit Milan! #M2L #MachineLearning #Research #Milan #SummerSchool #AI #ML

Gemma 2 is out!! Check out our latest model 😊

Please help us keep the school running! Any help, regardless of the amount, makes the difference!! 🙏🙏

Lovely spending some time chatting with participants, speakers and sponsors yesterday, in one of the most beautiful locations in Milan! #ML #M2LSummerSchool #tranvai

M2L talk sharing day has arrived 💥🎉🎊 Learn from the best of #M2L Summer School (Milan 🇮🇹) with free introductory talks on #DeepLearning & more. Day 1 lectures by Dilara Gokay, Mihaela Rosca, Lucas Beyer (bl16) and Armando Bellante. youtube.com/playlist?list=…

And the M2L Day 3 talks are now available! Check out the newest set of #M2L talks to learn what’s new in the world of LLMs, mixture of experts and theorem proving! Day 3 lectures by Thomas Mesnard, Diego de las Casas and Dr. Emily First can be found here youtube.com/playlist?list=…

And finally all the recordings are out! 🎦 I'm really grateful to all the speakers for yet another outstanding edition of M2L!! 🙏🤩✨

Gemma 3 270M (8-bit) at ~200 tk/s on iPad Air M3 with MLX

Google has released a new open source model... That runs on just 0.5 GB of RAM. Yes. You can fine-tune it for free to make it better than the giant models at your tasks. Quick steps to fine-tune Gemma 3 270M below

Gemma 270M can be finetuned and deployed in production *in minutes*!