🔥🎓New FastAI's free online course on NLP

It is called «A Code-First Introduction to Natural Language Processing». All code & videos are available for free online, make sure you save this link into bookmarks and at least review the content, because it provides opportunity not only to learn new skills, but to actually understand how NLP works.

Link: https://www.fast.ai/2019/07/08/fastai-nlp/

#NLP #NLU #DL #MOOC #FastAI #course

ODS breakfast in Paris! See you this Saturday at 10:30 at Malongo Café, 50 Rue Saint-André des Arts.

BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs

Authors claim really fast face detection on mobile devices.

Link: https://arxiv.org/abs/1907.05047

#FaceRecognition #CV #mobile #DL

Facebook, Carnegie Mellon build first AI that beats pros in 6-player poker

Main theoretical output: practical proof that carefully constructed AI algorithm can reach superhuman performance outside of two-player zero-sum games.

Training time: 8 days
Server: 64 core and 512 GB of RAM
Est. Cost to train: $150

The question whether that means end of the online poker, remains open for everyone to answer (or even try to train such bot themselves and take part).

Vote 🃏 if you believe the industry won't notice.
Vote 🤖 if you believe that it will be affected.

Link: https://ai.facebook.com/blog/pluribus-first-ai-to-beat-pros-in-6-player-poker/

Alan Turing will become a face of new £50 note

That's a great acknowledgment of the man who stands behind most of the theoretical computing.

Link: https://www.bbc.com/news/business-48962557
Most famous Turing's work 'On computable numbers': https://www.cs.virginia.edu/~robins/Turing_Paper_1936.pdf
Turing machine: https://en.wikipedia.org/wiki/Turing_machine

#Turing #Theory #Math #history

Useful and practical post on pandas indexing

Link: https://www.shanelynn.ie/select-pandas-dataframe-rows-and-columns-using-iloc-loc-and-ix/

#beginner #pandas #practical #novice #entrylevel

ODS breakfast in Berlin! See you this Wednesday at 08:30-10:00 at Einstein (Alexanderplatz 3, 10178 Berlin)
ODS Frühstück in Berlin! Wir sehen uns an diesem Mittwoch um 08:30 - 10:00 Uhr in Einstein cafe (Alexanderplatz 3, 10178 Berlin)

Large Memory Layers with Product Keys

Exploration of transformer architecture with special memory layer.

Link: https://arxiv.org/abs/1907.05242

#NLP #DL #transformer

Prisma founders launch Capture.

New app allows to take a photo of anything (even series and books) and get discuss it with random people online. All the processing is done on device. iOS only until autumn.

Link: https://techcrunch.com/2019/07/16/social-chat-app-capture-launches-to-take-a-shot-at-less-viral-success/
AppStore: https://apple.co/2xWnUUl

#Prisma #DL #CV #product

​​Generative Modeling by Estimating Gradients of the Data Distribution

Paper on a different approach to generative modeling. We can estimate gradients of the data distribution and sample with Langevin dynamics. No adversarial method and no approximation for tractable training. Record-breaking inception score of 8.91 on CIFAR-10.

Github: https://github.com/ermongroup/ncsn
ArXiV: https://arxiv.org/abs/1907.05600

#GAN #CIFAR #cv #dl

We count every opinion and listen to your feedback, so please vote.

We also preparing special event for the chat creation, so stay tuned for the announcement

Microsoft open-sourced scripts and notebooks to pre-train and finetune BERT natural language model with domain-specific texts

Github: https://github.com/microsoft/AzureML-BERT
Earlier: https://yangx.top/opendatascience/837

#Bert #Microsoft #NLP #dl

Checkout our friends' recent publication.

ODS breakfast in Paris! See you this Saturday at 10:30 at Malongo Café, 50 Rue Saint-André des Arts.

​​TabNine showed deep learning code autocomplete tool based on GPT-2 architecture.

Video demonstrates the concept. Hopefully, it will allow us to write code with less bugs, not more.

Link: https://tabnine.com/blog/deep
Something relatively similar by Microsoft: https://visualstudio.microsoft.com/ru/services/intellicode

#GPT2 #TabNine #autocomplete #product #NLP #NLU #codegeneration

Great collection of practical rules for routine DS engineering / research job.

Machine Learning in a company is 10% Data Science & 90% other challenges, this pdf provides a great deal of principals and solutions to deal with them.

We can only recommend saving this post to your Saved Messages by forwarding it to yourself.

Link: http://martin.zinkevich.org/rules_of_ml/rules_of_ml.pdf

#cheatsheet #advice #practical #common #shouldbesaved

​​YouTokenToMe, new tool for text tokenisation from VK team

Meet new enhanced tokenisation tool on steroids. Works 7-10 times faster alphabetic languages and 40 to 50 times faster on logographic languages, than alternatives.

Under the hood (watch source) there is C++ implementation with python bindings, using Byte Pair Encoding (BPE) algorithm. YouTokenToMe beats #SentencePiece by Google and #fastBPE, created by a researcher from Facebook AI Research in terms of speed.

Github: https://github.com/vkcom/YouTokenToMe
Medium: https://medium.com/@vktech/youtokentome-a-tool-for-quick-text-tokenization-from-the-vk-team-aa6341215c5a
Byte Pair Encoding: https://arxiv.org/abs/1508.07909

This improvement for everyday used toolset deserves minimum 50 claps at Medium and a Star on github!
Let’s give VK research team appreciation from the community they deserve 👏!!

What’s wrong with transformer architecture: an overview

How the Transformers broke NLP leaderboards and why that can be bad for industry.

Link: https://hackingsemantics.xyz/2019/leaderboards/

#NLP #overview #transformer #BERT #XLNet