Computer Vision news from RSIP VISION. April 2019

Automatic feature selection:

EGU is still ongoing, but I am burning to share with you some of my findings already!

Research group in University of Lausanne developed a pretty promising algorithm for automatic feature selection based on General Regression Neural Network (GRNN, also known as Nadaraya-Watson Estimator). The idea behind is pretty simple and therefor powerful - why won't we build the simplest network that can train really fast and brute force all possible combination of features to check how they affect accuracy, learning rate etc and than select the best performing once.

Here is Python implementation on GitHub: https://github.com/federhub/pyGRNN
Also, check their poster: https://github.com/federhub/pyGRNN/blob/master/EGU2019_FS_using_simple_and_efficient_ML_models.pdf

Stay tuned, subscribe and share!
xoxo

​​Autodesk claims to use GANs to design a chair.

First-of-its-kind chair from Philippe Starck and Kartell. Imagined by a human and cocreated with intelligent generative design

Link: https://adsknews.autodesk.com/news/starck-intelligent-generative-design
Generative design explanation link: https://www.autodesk.com/solutions/generative-design

#Autodesk #generativedesign

What does the future data scientist look like?
http://bit.ly/2IgHW3g

Explaining Random Forest (with Python Implementation)
http://bit.ly/2X2F6SI

The Growing Trend of Data Science in Japan
http://bit.ly/2WWRIee

Let's get LEGO'd! (in R)
http://bit.ly/2IdGaQw

Neural Machine Translation With Attention Mechanism
http://bit.ly/2IcvjpN

The first AI-generated textbook shows what robot writers are actually good at
http://bit.ly/2X8zp5O

Aroma: Using machine learning for code recommendation
http://bit.ly/2IkRzy1

doccano - open source text annotation tool for ML practitioner
http://bit.ly/2XdJFtp

Introducing d3-regression
http://bit.ly/2IihiqH

7 Qualities Your Big Data Visualization Tools Absolutely Must Have and 10 Tools That Have Them
http://bit.ly/2IlIEfv

The insideBIGDATA IMPACT 50 List for Q2 2019
http://bit.ly/2XdjRO2

Mathematicians Discover the Perfect Way to Multiply
http://bit.ly/2XaL8AM

dabl, the Data Analysis Baseline Library
http://bit.ly/2IkKvBs

Top 5 Interesting Applications of GANs for Every Machine Learning Enthusiast!
http://bit.ly/2XhEY25

A too complicated map
http://bit.ly/2Ikm9rt

The EU commission introduces guidelines for achieving a ‘Trustworthy AI’
http://bit.ly/2IhXILj

Best of arXiv.org for AI, Machine Learning, and Deep Learning – March 2019
http://bit.ly/2XbZjWe

Another all cheat-sheets in one place
http://bit.ly/2Il6eZO

Starting Your Career in Data Science: What Are Your Options?
http://bit.ly/2Ikrjn8

Fusing data and design to supercharge innovation—in products and processes
https://mck.co/2X58uI2

🔥Singing voice conversion system developed at FAIR-Tel Aviv.

This can transform someone's singing voice into someone else's voice.

YouTube: https://www.youtube.com/watch?v=IEpkGenLnjw
Link: https://venturebeat.com/2019/04/16/facebooks-ai-can-convert-one-singers-voice-into-another/
ArXiV: https://arxiv.org/abs/1904.06590

#voiceconversion #audiolearning #DL #Facebook

Boom! Weekly selection from DSB. Issue #50
The TOP 10 most popular links for the week

1. Here's who will die on 'Game of Thrones' according to AI
http://bit.ly/2IjaKIm

2. Another 10 Free Must-See Courses for Machine Learning and Data Science
http://bit.ly/2X5kTfc

3. The first AI-generated textbook shows what robot writers are actually good at
http://bit.ly/2X8zp5O

4. Mathematicians Discover the Perfect Way to Multiply
http://bit.ly/2XaL8AM

5. Google launches an end-to-end AI platform
https://tcrn.ch/2XeJnCR

6. Another all cheat-sheets in one place
http://bit.ly/2Il6eZO

7. Serverless for data scientists
http://bit.ly/2Ijg12B

8. A too complicated map
http://bit.ly/2Ikm9rt

9. How to Version Control Jupyter Notebooks
http://bit.ly/2XdWUL0

10. Best of arXiv.org for AI, Machine Learning, and Deep Learning – March 2019
http://bit.ly/2XbZjWe

Google’s Best AI Just Flunked a High School Math Test
http://bit.ly/2XyqGdg

How and Where to Find Great Data Science Jobs
http://bit.ly/2IxDOfx

A Hands-On Introduction to Deep Q-Learning using OpenAI Gym in Python
http://bit.ly/2XwQp60

Forecasting: Principles and Practice
http://bit.ly/2IxTIGv

Timeline of Mathematics
http://bit.ly/2IEmDsP

WIPO Launches AI-Based Image Search Tool for Brands
http://bit.ly/2IzYTWo

Debunking 10 Misconceptions About AI
http://bit.ly/2IA0lbk

Actively curated list of awesome BI tools.
http://bit.ly/2XBbaxr

Advanced NLP with spaCy
http://bit.ly/2IAqS8K

Traffic simulation
http://bit.ly/2XwWgbu

A Recipe for Training Neural Networks by Andrej Karpathy

New article written by Andrej Karpathy distilling a bunch of useful heuristics for training neural nets. The post is full of real-world knowledge and how-to details that are not taught in books and often take endless hours to learn the hard way.

Link: https://karpathy.github.io/2019/04/25/recipe/

#tipsandtricks #karpathy #tutorial #nn #ml #dl

OpenAI’s MuseNet architecture to generate music.

#MuseNet — neural network which discovered how to generate music from first 5 or so notes, using many different instruments and styles.

Post: https://openai.com/blog/musenet/

MuseNet will play an experimental concert today from 12–3pm PT on livestream: http://twitch.tv/openai

#audiolearning #musicgeneration #OpenAI #soundgeneration

Deep Learning lecture
The full deck of (600+) slides, by Professor Gilles Louppe. PDF file available here:

Deep Learning lecture

Boom! What to read on the weekend, Issue #60

Exploring data to form better questions
http://bit.ly/2XBdTXH

AI is helping old video games look like new
http://bit.ly/2IClMsq

How to Train Your OpenAI Five
http://bit.ly/2XBdW5P

What Machine Learning needs from Hardware
http://bit.ly/2IAupnw

Why a data scientist is not a data engineer
https://oreil.ly/2IvjhrK

The Coming AI Autumn
http://bit.ly/2XAekBl

Artificial Intelligence for Pharmaceutical Research and Development
http://bit.ly/2IyM7rl

How to create your own R package with RStudio
http://bit.ly/2IyMuSL

Tutorials on getting started with PyTorch and TorchText for sentiment analysis.
http://bit.ly/2XyOiyE

Training Deep Convolutional GANs to generate Anime Characters
http://bit.ly/2Iy5Pn6

Nitro PDF Pro v12.12.1.522 32bit 64bit
Activated.

Everything you need for PDF documents. Create, convert, edit, combine, protect, sign, share, OCR and more.

Boom! Weekly selection from DSB. Issue #51
The TOP 10 most popular links for the week

1. 10 free Data Science books you must read
http://bit.ly/2XxpgQm

2. Curated list of Python resources for data science.
http://bit.ly/2IxFXI7

3. How Algorithms Know What You’ll Type Next
http://bit.ly/2IxVRlx

4. 8 Useful R Packages for Data Science You Aren’t Using (But Should!)
http://bit.ly/2Ix6VPV

5. Google’s Best AI Just Flunked a High School Math Test
http://bit.ly/2XyqGdg

6. AI developed a whole new sport
https://engt.co/2XC46AJ

7. Forecasting: Principles and Practice
http://bit.ly/2IxTIGv

8. A simple introduction to the basics of data science
http://bit.ly/2ICj3PI

9. One Model to Rule Them All
http://bit.ly/2IEr5Yz

10. Timeline of Mathematics
http://bit.ly/2IEmDsP

Machine Learning Workflows

Pretty articel about how machine learning / AI works in the organization.

Link: https://skymind.ai/wiki/machine-learning-workflow

#management #production #workflow