A few neat things in the next TF release:

- The Keras mixed precision API moves to stable
- Multi-worker mirrored distribution support moves to stable
- Experimental support for parameter server distribution in Keras
- Full NumPy API implementation (tf.experimental.numpy)
Using mixed precision: tensorflow.org/guide/mixed_pr…
Using parameter server distribution: tensorflow.org/tutorials/dist…
Learn more about the NumPy API in TensorFlow: tensorflow.org/guide/tf_numpy

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with François Chollet

François Chollet Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @fchollet

15 Nov
The territory, the map, and the cartographer each belong in entirely different categories. Yet, we tend to confuse them with each other

The thing, the abstraction that captures an actionable aspect of the thing, and the intelligence that is capable of generating new abstractions
In particular, don't confuse a model and the ability to produce new models. Humans excel at operationalizing the abstractions they invent into highly effective machines, but so far our machines have close to zero ability to generate abstractions of their own
A road takes you from A to B, but a road-building company takes you from anywhere to anywhere else.

An operationalized abstraction solves the task for which it was designed, but intelligence -- the ability to produce abstractions -- can solve arbitrary tasks
Read 4 tweets
11 Nov
In a majority-democrat country, nearly 100% of top performing FB content consists of far-right propaganda and conspiracies.

Is it because only hardcore right wingers use FB anymore, or is it because FB recommendations amplify far-right propaganda at the expense of all else?
Do Trump-supporting FB board members and executives have a say in what their algorithm promotes? In fact, they do.

I don't think History will remember someone like Lecun as a convnet pioneer. He will be remembered as the man who was in charge of AI at Facebook during the decade when Facebook's algorithms weakened democracy worldwide and triggered a global wave of far-right populism.
Read 6 tweets
8 Nov
Empathy is good. Calls for unity are good. But let's stop pretending that the "two sides" are symmetrical and morally equivalent. Stop saying, "oh, you think X is bad, but the other side just thinks the same about you".

Along every dimension, the picture is incredibly lopsided.
Maybe the most glaring asymmetry is what was at stake in this election. Biden won, and Trump supporters have nothing to fear from this result. In policy terms, they'll actually end up better off.

Had Trump won, many people would have definitely not been ok. Lives were at stake.
And no, CNN & NYT are not equivalent to Fox News.

Conspiracy theories against Biden's family are not equivalent to Trump's actual corruption.

Antiracism activists are not equivalent to far-right militias.

Life-saving medical facts are not equivalent to life-endangering lies.
Read 5 tweets
6 Nov
If you plot the linear trendlines for the GA and PA data (very linear in the case of PA) you see Biden winning both. Though not by much
GA: check. PA in the morning.
The emotional rollercoaster on election night would have been very different if all ballots had been counted quickly.
Read 4 tweets
4 Nov
Plenty of ballots left to count (and remember, things will get bluer over time as mail-in ballots get tallied). But I hoped we'd see a clear winner tonight, and the fact that it's looking unlikely is disappointing beyond words
Time and time again, I keep believing in everything good about America, and America keeps breaking my heart.

Like in 2016, I'm taken by surprise. I am fundamentally unable to foresee these things, because it would mean having such a low view of America.
It's like if your best friend turns out to be a murderer. You couldn't foresee it -- you wouldn't have been their friend if you could.

I just can't come to terms with the awfulness. So I just keep getting my heart broken with every news
Read 4 tweets
24 Oct
It's easy to use deep learning to generate notes that sound like music, in the same way that it's easy to generate text that looks like natural language.

But it's nearly impossible to generate *good* music that way, much like you can't generate a good 2-page story or poem
With two caveats:

1. Plagiarism. If you near-copy large chunks of a good piece, these chunks will be good.

2. Large-scale curation. If you generate thousands of samples and hand-pick the best, they may be good by happenstance (especially for music, where the space is smaller)
However, algorithms (and ML in particular) absolutely do have a role to play in music creation. What's broken is the general approach of statistical mimicry, e.g. raw deep learning.

To generate good music programmatically, you need an algorithmic model of what makes music good.
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!