If you haven't already filled in the questionnaire, please complete it.

Now we are going to adjust the channel policy according to the responses, so opinion of 80 people will set it.

You can submit your responses to make YOUR opinion matter.

We will try to follow the requests of the auditory, but we need your responses.

3 min to complete: https://goo.gl/forms/VBYApzVRGCUhzb713

🔥🚀UPD2: Thank you! +50 responses in less than an hour!

UPD : 🔥💪135 people are making the future of this channel / 1545 views of the post with link.

And there are a lot of words of support, thank you, dear subscribers!

This is a day to remembered. #OpenAI 's team of five neural networks, OpenAI Five, has started to defeat amateur human teams (including a semi-pro team) at Dota 2:
https://blog.openai.com/openai-five/

It is important, because Dota2 is a way more complicated game than Chess or Go, where #AI has already surpassed human players.

#rl #reinforcementlearning #dl #dota2 #lstm

Tensorflow: The Confusing Parts (1)

The tutorial for beginners by Jacob, Google AI Resident. This can be nice intro for those, who wanted to get familiar with #TF

This is thorough introduction to the concepts underlying Tensorflow’s API; such as nodes, graphs and sessions.

https://jacobbuckman.com/post/tensorflow-the-confusing-parts-1/?utm_source=telegram&utm_medium=opendatascience

#tensorflow #tutorial #novice #beginner

Thank you all, folks, for support and suggestions, which you had sent through the questionnaire form.

You submitted 252 responses and our team now is able to know you better.

Most of you approve posting frequency and the content. We will continue to maintain the channel, especially after such warm approval!

Thank you, fellow Data Scientists.

Face recognition is now available as a JS-package with the help of face-api.js. It is built on top of #Tensorflow (js version).

https://itnext.io/face-api-js-javascript-api-for-face-recognition-in-the-browser-with-tensorflow-js-bcc2a6c4cf07?gi=a277ad002e2a

#cv #js #tf #dl #cnn

Automated Deep Reinforcement Learning Environment for Hardware of a Modular Legged Robot, from (sic!) Disney Research. They created configurable robots with an "env.reset()" that works on the real hardware enabling popular typical Gym-style ML directly on robots!

Link: https://www.disneyresearch.com/publication/automated-deep-reinforcement-learning-environment-for-hardware-of-a-modular-legged-robot/

#robo #rl #disney

A visual introduction to machine learning.

It is an interactive website, which would be really useful to the beginners, as a perfect visual explanation of how decision trees work. It shows how one can go from statistical parametric evaluation to decision tree building.

Link: http://www.r2d3.us/visual-intro-to-machine-learning-part-1/?utm_source=telegram&utm_medium=opendatascience

#decisiontrees #beginner #novice #firststep #howitworks

Model Tuning and the Bias-Variance Tradeoff
(part II of Visual Introduction to Machine Learning by r2d3)

Bias-Variance tradeoff happens because you have to find optimal balance between model being too simple and too complex. Too complex models tend to overfit — to become to adapted to the training data, so the results on the testing (new, unknown to model) data become less accurate. The article explains with the example from previous part how this actually works.

http://www.r2d3.us/visual-intro-to-machine-learning-part-2/?utm_source=telegram&utm_medium=opendatascience

#decisiontrees #beginner #novice #firststep #howitworks

Post on the estimation of the likelihood of the ads closing due to actually selling the good, which was the goal of Avito’s #Kaggle contest. Author built a #neuralnetwork solution and had 18th place.

https://towardsdatascience.com/kaggle-avito-demand-challenge-18th-place-solution-neural-network-ac19efd6e183

Nice hot start on data streaming with #Spark

https://towardsdatascience.com/learn-how-to-code-and-deploy-machine-learning-models-on-structured-streaming-564749840516

#scala #bigdata #streaming #kafka #tutorial #novice #beginner

Practical Advice for Building Deep Neural Networks

Some practical tips for training deep neural networks based the experiences (rooted mainly in TensorFlow). Some of the suggestions may seem obvious, but they weren’t at some point. Other suggestions may not apply or might even be bad advice for particular task: use discretion!

https://pcc.cs.byu.edu/2017/10/02/practical-advice-for-building-deep-neural-networks/

#neuralnetworks #dl #tensorflow

On the Adam optimizer convergence

Investigation of the convergence of popular optimization algorithms like Adam, RMSProp and propose new variants of these methods which provably converge to optimal solution in convex settings.

Link: https://openreview.net/forum?id=ryQu7f-RZ
PDF: https://openreview.net/pdf?id=ryQu7f-RZ

#iclr2018 #neuralnetworks #optimizers

Pitfalls of Batch Norm in TensorFlow and Sanity Checks for Training Networks

Some more practical advices on #tensorflow training with source code and reference links to look up.

https://medium.com/@theshank/pitfalls-of-batch-norm-in-tensorflow-and-sanity-checks-for-training-networks-e86c207548c8

#beginner #novice #dl #tutorial

Introduction to Python Ensembles

Ensemble is an approach for mixing algorithms to obtain strong sides of all the approaches.

https://www.kdnuggets.com/2018/02/introduction-python-ensembles.html

#ensemble #stacking #tutorial #beginnig #novice

Adversarial attack — type of input or a mask applied to the input of the machine learning model to make it wrong. It is a way to cheat with the output, to ‘fool’ the algorithm.

«Attacking Machine Learning with Adversarial Examples» at Open AI blog covers the basics and provides some examples.

Open AI blog article: https://blog.openai.com/adversarial-example-research/

#adversarialattack #openai #novice #beginner

New attack on neural networks can alter the purpose of the neural network.

A surprising adversarial attack, whereby a perturbation to all input images can "reprogram" a poorly-defended neural network to change its task entirely. e.g. turn an ImageNet classifier into a network that counts squares.

Arxiv: https://arxiv.org/pdf/1806.11146.pdf

#Goodfellow #gbrain #adversarialattack

Most common pitfalls, you can encounter when training neural network.

http://telegra.ph/Most-common-neural-network-mistakes-07-01

#beginner #novice #dl #tutorial

Deep Mind announced that its agent beated human performance in Quake III CTF (Capture The Flag)

https://deepmind.com/blog/capture-the-flag/

#rl #quake3 #deepmind

ModaNet: A Large-Scale Street Fashion Dataset with Polygon Annotations

Latest segmentation and detection approaches (DeepLabV3+, FasterRCNN) applied to street fashion images. Arxiv paper contains information about both: net and dataset.

Arxiv link: https://arxiv.org/abs/1807.01394
Paperdoll dataset: http://vision.is.tohoku.ac.jp/~kyamagu/research/paperdoll/

#segmentation #dataset #fashion #sv

Hey, our fellow colleagues at OpenDataScience community are labeling a meme dataset. You can help them with the markup just by viewing memes in this bot: @MemezoidBot

#DataSet #labeling