Issue 44

In the News

A look at IBM’s Watson 5 years after its debut

5 years ago, Watson became known worldwide by beating human champions at Jeopardy. Since then, IBM has been using it as an umbrella name to cover most of its ML / AI products, leading to some controversies. How far have they actually gotten?

AI’s Language Problem

Machines that truly understand language would be incredibly useful. But we don’t know how to build them.

Twitter facial analysis reveals demographics of presidential campaign followers

If you follow Hillary Clinton or Donald Trump on Twitter, your face has probably been analyzed by a machine to determine your age, ethnicity, and social influence.

Also in the news this week...

Uber and Volvo announce $300M joint venture to develop self driving cars. First fleet to arrive in Pittsburgh soon. More

Learning

Analyzing over 1M hotel reviews

How to use sentiment analysis to crunch millions of TripAdvisor reviews and understand how people feel about hotels in different cities

Image Completion with Deep Learning in TensorFlow

Fantastic post by Brandon Amos on how to achieve image completion (i.e. fill missing parts in images) with Generative Adversarial Networks. Great for beginners as well as more advanced coders, explains the theory and the code.

AMA with the Google Brain team

Interesting discussion on Reddit with a few of Google Brain team's members. Questions on their current work and links with Deepmind, their hardware innovations (TPUs) and how they differ from academic labs in their daily work and objectives.

Deep Learning in Fashion

Tutorial on clothing matching

Software tools & code

Top Deep Learning projects

This is a list of the most popular Deep Learning projects on Github. The top 2 are (of course) Tensorflow and Caffe. But you may find a few hidden gems lower in the list.

Loosing weight with ML

Interesting endeavor where Ariel Faigon tracked his lifestyle for a few months and then used ML to determine how to losse weight. He shares insights and code. The approach seems to have worked. If you are curious, here are the lifestyle choices that contributed most to his weight loss (according to ML) => stay at home, sleep, bacon, egg, parmesan and olive oil. Surprising!

Machine Learning exercises

Long list of machine learning exercises in Python, taken from Andrew Ng's class on Coursera. Many topics are covered, from linear regression to k-means clustering, SVMs and neural networks.

Bayesian Sentiment Classifier

This project (along with code) explores a few optimizations to typical Naive Bayesian methods for sentiment analysis.

Hardware

Samsung turns IBM's brain-like chip into a digital eye

Samsung is using IBM's 'neural network on a chip' TrueNorth tech for vision image processing - demoing gesture recognition on a TV

Titan X benchmark

Benchmarking the new Nvidia Titan X by querying 1.1 billion taxi trips made in New York City over the course of six years.

Intel announces Knights Mill

Intel wants a seat in the Deep Learning market - announces a Xeon Phi For Deep Learning

Some thoughts

Dreaming of names with RBMs

Dreaming of human names, geographic names and Github repository names. Serves as an introduction to RBMs.

Neural Net turns NYC into moving painting

About

This newsletter is a collection of AI news and resources curated by @dlissmyr. If you find it worthwhile, please forward to your friends and colleagues, or share on your favorite network!

Share on Twitter · Share on Linkedin · Share on Google+

Suggestions or comments are more than welcome, just reply to this email.

Thanks!