​Facebook has released #PyText — new framework on top of #PyTorch.

This framework is build to make it easier for developers to build #NLP models.

Link: https://code.fb.com/ai-research/pytext-open-source-nlp-framework/

🔗 Open-sourcing PyText for faster NLP development
We are open-sourcing PyText, a framework for natural language processing. PyText is built on PyTorch and it makes it faster and easier to build deep learning models for NLP.

​New book by Andrew Ng

Drawn from his experience leading Google Brain, Baidu's AI Group, and Landing AI, this 5-step Playbook provides a roadmap for your company to transform into a great AI company.

Site: https://landing.ai/ai-transformation-playbook
Direct link: https://d6hi0znd7umn4.cloudfront.net/content/uploads/2018/12/AI-Transformation-Playbook.pdf

#book

🔗 AI Transformation Playbook How to lead your company into the AI era - Landing AI

​Трудности создания Open Source Machine Learning библиотеки на примере Apache Ignite ML

🔗 Трудности создания Open Source Machine Learning библиотеки на примере Apache Ignite ML
Доклад для тех кому интересно машинное обучение, написание библиотек для open source, распределенные вычисления и задачи, которые приходится решать в ходе ра...

​How to Improve Deep Learning Model Robustness by Adding Noise

https://machinelearningmastery.com/how-to-improve-deep-learning-model-robustness-by-adding-noise/

🔗 How to Improve Deep Learning Model Robustness by Adding Noise
Adding noise to an underconstrained neural network model with a small training dataset can have a regularizing effect and reduce overfitting. Keras supports the addition of Gaussian noise via a separate layer called the GaussianNoise layer. This layer can be used to add noise to an existing model. In this tutorial, you will discover how …

🎥 Онлайн мастер-класс от Паулы Янушкевич (CQURE) по информационной безопасности
👁 4 раз ⏳ 9922 сек.
Запись мастер-класса без перевода, на английском языке. Доступны русские субтитры.

​Link: https://www.technologyreview.com/s/612561/a-radical-new-neural-network-design-could-overcome-big-challenges-in-ai/

🔗 A radical new neural network design could overcome big challenges in AI
Researchers borrowed equations from calculus to redesign the core machinery of deep learning so it can model continuous processes like changes in health.

Object Detection using Google AI Open Images
https://towardsdatascience.com/object-detection-using-google-ai-open-images-4c908cad4a54?source=collection_home---4------0---------------------

​BigGANs: AI-Based High-Fidelity Image Synthesis

🔗 BigGANs: AI-Based High-Fidelity Image Synthesis
This episode was supported by insilico.com. "Anything outside life extension is a complete waste of time". See their papers:
- Papers: https://www.ncbi.nlm.nih.gov/pubmed/?term=Zhavoronkov%2Ba
- Website: https://insilico.com/

The paper "Large Scale GAN Training for High Fidelity Natural Image Synthesis" is available here:
- Paper: https://arxiv.org/abs/1809.11096
- Try it here: https://colab.research.google.com/github/tensorflow/hub/blob/master/examples/colab/biggan_generation_with_tf_hub.ipynb

We would li

​Quick poll, please fill where are you from: https://goo.gl/forms/AacYr2lDzxBg1vpf2

Poll is available until GMT 23:59. Please, vote, this will affect content as well as other activities, which can be created for our community.

🔗 Quick @opendatascience poll

​Why Deep Learning Works: ICSI UC Berkeley 2018

🔗 Why Deep Learning Works: ICSI UC Berkeley 2018
Recent development in the Theory of Heavy Tailed Self Regularization for Deep Neural Networks. An invited talk at The International Computer Science Institut...

Music Genre Classification with Python
https://towardsdatascience.com/music-genre-classification-with-python-c714d032f0d8?source=collection_home---4------1---------------------

​Машинное обучение доп главы 13.Рекомендательные системы − 2

🔗 Машинное обучение доп главы 13.Рекомендательные системы − 2
08.12.2018

у кого есть книга "прикладное машинное обучение с помощью scikit-learn и tensorflow"? поделитесь, пожалуйста

​🔗 TensorSpace.js
Home page of TensorSpace.js

Processing Time Series Data in Real-Time with InfluxDB and Structured Streaming
https://towardsdatascience.com/processing-time-series-data-in-real-time-with-influxdb-and-structured-streaming-d1864154cf8b?source=collection_home---4------0---------------------

What’s the fuss about Regularization?
https://towardsdatascience.com/whats-the-fuss-about-regularization-24a4a1eadb1?source=collection_home---4------2---------------------

​Jax is NumPy on CPU and GPU with automatic differentiation and JIT compilation.

Link: https://github.com/google/jax

🔗 google/jax
GPU- and TPU-backed NumPy with differentiation and JIT compilation. - google/jax

​Using Kubernetes to Offer Scalable Deep Learning on Alibaba Cloud - Kai Zhang & Yang Che, Alibaba

🔗 Using Kubernetes to Offer Scalable Deep Learning on Alibaba Cloud - Kai Zhang & Yang Che, Alibaba
Using Kubernetes to Offer Scalable Deep Learning on Alibaba Cloud - Kai Zhang & Yang Che, Alibaba Running deep learning (DL) jobs requires end to end workflo...

​Deep Learning: What is it good for? - Prof. Ankit Patel - Rice University

🔗 Deep Learning: What is it good for? - Prof. Ankit Patel - Rice University
"In this talk, we will introduce deep learning and review some of the key advances in the field focusing on current attempts at a theoretical understanding. ...

​MIT AI: Deep Reinforcement Learning (Pieter Abbeel)

🔗 MIT AI: Deep Reinforcement Learning (Pieter Abbeel)
Pieter Abbeel is a professor at UC Berkeley, director of the Berkeley Robot Learning Lab, and is one of the top researchers in the world working on how to ma...

​Physicists have discovered what makes neural networks so extraordinarily powerful

🔗 Physicists have discovered what makes neural networks so extraordinarily powerful
Nobody understands why deep neural networks are so good at solving complex problems. Now physicists say the secret is buried in the laws of physics.