We are so excited to be joining Tensorflow as TensorFlow.js, an ecosystem of tools and libraries for machine learning in JavaScript.
Find out more at https://t.co/8tjLM7TKGO.
We’re looking forward to building the next chapter of ML in JavaScript with you!
#TFDevSummit kicks off tomorrow with the Keynote at 9:30 AM PDT!
Tune in for a full day of highly technical talks, demos, and conversation with the TensorFlow team and community
Watch the livestream here → https://t.co/kMcvHRNIS0
Tune in to the TensorFlow developer summit tomorrow, March 30th at 10:30AM PDT to see @dsmilkov and @nsthorat give an update on deeplearn.js!
Livestream URL available at: https://t.co/kFUDxibEEC
Our work on "World Models" is out!🌐
Can neural network agents dream, and learn inside of their own dream worlds?
Read more to find out:
Full Interactive Article: https://t.co/diRMWBqpys
PDF: https://t.co/YZ9Wehzyfv
Me: If only there was a nice write up and starter code for all this amazing Music+ML+Web coming out this week ...
@tBlankensmith (two desks away): here you go!
https://t.co/QrwaxcSiG7
Check out MusicVAE, project Magenta’s #machinelearning model that allows you to blend and explore musical scores. Additionally, there are 3 browser-based apps to play with, a MusicVAE package built on @deeplearnjs, along with several pre-trained models → https://t.co/dwuMwckI7T.
Check out Magenta's newest model, called MusicVAE. Hear samples, generate your own music in a Colab, and learn to use their @deeplearnjs implementation, all at https://t.co/ICps6ka0IR!
Here's a video, too: https://t.co/64l8B19utM
Lots of exciting work to share today about #MusicVAE on the #Magenta blog: a full description, a JavaScript library built on @deeplearnjs, 3 browser-based apps to try it yourself, and examples galore!
https://t.co/mrfOp9rdu6
Play around with @OpenAI 's new Reptile meta-learning algorithm to do few-shot learning in your browser using @deeplearnjs and @observablehq. Crazy how easy it is to implement new #DeepLearning papers in #javascript now. https://t.co/e28Q3RHyLA
New pen, The Neural Drum Machine!
🥁 Try it here: https://t.co/3Lgm0EABt5
This one uses the "Drums RNN" model from @TensorFlow Magenta to generate drum patterns. And, once again, @deeplearnjs + Tone.
A neural network w/ 1 hidden layer (of infinite width) can recreate any function. Explore this property in real life by drawing your own data and using @deeplearnjs to fit a neural-net to it in an @observablehq notebook. #DeepLearning https://t.co/E2Avito9Dq
I just published a gentle @observablehq introduction to @deeplearnjs!
Along the way, the quirks of SGD became obvious since I could quickly run "experiments" by changing hyperparameters via UI controls.
Data science on the web has a bright future!
https://t.co/ZvK208WBys