Christian Rehm

AI @pipedrive @1kxnetwork. Moving fast, breaking things.

[Christian Rehm @Wadenschwinger]

Listened to some very interesting talks from @spotify @hopsworks @twitter @salesforce at #FeatureStoreSummit last night. Thanks for organizing @jim_dowling and team, looking forward to day 2!

[Christian Rehm @Wadenschwinger]

@darachennis Did you use Internet Explorer to tweet this?

[Christian Rehm @Wadenschwinger]

We have just released a blog post on our current project implementing a BERT based #NLP model for multilabel text classification at @Wayfair @WayfairTech @DataWayfair
https://t.co/OmVi3jqo7l

[Christian Rehm @Wadenschwinger]

@_inesmontani donezo!

[Christian Rehm @Wadenschwinger]

@fchollet @ForckNiklas

[Christian Rehm @Wadenschwinger]

@heinz_gies @darachennis Will do! I hope it's a bit less furious than the one in Berlin though :)

[Christian Rehm @Wadenschwinger]

Great talk today by @seb_ruder on cross-lingual and continual transfer learning in NLP @DeepMindAI

[Christian Rehm @Wadenschwinger]

NLP Today 😄 @yoavgo @spacy_io #spaCyIRL

[Christian Rehm @Wadenschwinger]

@darachennis @heinz_gies @DPC_22 revolutionary stream processing aus Berlin? 😁

[Christian Rehm @Wadenschwinger]

@eturner303 @Shujian_Liu

[Jared Palmer @jaredpalmer]

When you merge your own pull request

[Christian Rehm @Wadenschwinger]

https://t.co/nANB9nkRgs 🍕

[Christian Rehm @Wadenschwinger]

Finally 😀

[TensorFlow @TensorFlow]

The TensorFlow Extended (TFX) Python 3 Release 0.13 is now available!

Find support and other goodies here → https://t.co/yuCi5LFoMv
#TFX #TensorFlow

[Christian Rehm @Wadenschwinger]

@heinz_gies

[Christian Rehm @Wadenschwinger]

@karpathy describing his process for applying neural networks to a new problem https://t.co/f7Fjqm1gE2

[Gene Kogan @genekogan]

SampleRNN generated death metal on 24/7 livestream. that there's no humans and it's infinite is the most metal thing ever 🎸💀���🤘 https://t.co/PL57CfD1pz

[Greg Brockman @gdb]

I asked @ilyasut how to set neural network init. He accidentally replied with a poem:

You want to be on the edge of chaos
Too small, and the init will be too stable, with vanishing gradients
Too large, and you'll be unstable, due to exploding gradients
You want to be on the edge

[Christian Rehm @Wadenschwinger]

How artwork personalization is done @netflix https://t.co/hzYyQIIPWp

[Gene Kogan @genekogan]

I always thought of Schmidhuber as an artist. His emphasis on authenticity, and willingness to write about unusual topics with creativity and rigor at the same time, e.g. https://t.co/35EwfXWX6S
Awards don't deserve him.