Here is the second half of my January 2017’s blog reads! This month has been a pretty successful month for AI and ML. Although 2017 may be a very harsh year for everyone with all the world problems happening around the world, hopefully research and technology advancement will keep striking.
One Dataset, Visualized 25 Ways I always love posts from Flowing Data and this post was especially wonderfully written. Description of this project: “It’s life expectancy data by country, it’s from the the World Health Organization and it spans 2000 to 2015. Each chart provides a different focus and interpretation”.
Big Companies News
“OK Facebook”—Why stop at assistants? Facebook has grander ambitions for modern AI After Apple’s Siri, Google’s Now, Amazon’s Alexa and Microsoft’s Cortana, it is time for Facebook to investe heavily in the production of a powerful AI, capable of conversing and answering the every day needs of humans.
Talking AI Disruption With the Man Who Built Google’s ‘Brain’ Of course, Facebook is not the only big name investing heavenly in the production of an optimal AI for its customer. In China, the Baidu team led by Andrew Ng is Little Fish.
The Chinese News App with 600 Million Users That You’ve Never Heard Of This new app called Toutiao uses artificial intelligence to curate headline recommendations. It has met amazing success with the Chinese users.
Learning the Theory
An executive’s guide to machine learning Here is an easy and fun read to understand the fundamentals of Machine Learning. The post talks about how the traditional industries can use ML and the importance of management into implanting ML into the business.
A short history of color theory Have you ever heard about color theory? Do you know what is a color tree? Well, now is your chance to learn more about the whole science behind color. No, it is not just 7 colors forming a rainbow.
- Scaling Recommendation Engine: 15,000 to 130M Users in 24 Months [RS Labs]
- Recognizing Traffic Lights With Deep Learning
- Machine-Learning Algorithm Identifies Tweets Sent Under the Influence of Alcohol
- Analyzing Emotions using Facial Expressions in Video with Microsoft AI and R
- Was 2016 especially dangerous for celebrities? An empirical analysis.
- Deep learning algorithm does as well as dermatologists in identifying skin cancer
- The United States Of Artificial Intelligence Startups
- The 2017 Voice Report by VoiceLabs
- Understanding How Machines Learn, Through Prototyping
- Lane Following Autopilot with Keras & Tensorflow
- MAKING IT BIG What three years and 75,000 shows in New York tell us about the chance your favorite band will succeed.
- Artificial intelligence predicts when heart will fail
AI Playing Poker January 2017
The biggest AI news of the month (and starting the 2017 year strong) is the new Poker playing AIs. Yes, there are two different projects being tested currently. The first AI was written by research teams from Canada (University of Alberta) and Czech Republic (Czech Technical University), called DeepStack. The AI can easily beat up many professionals from the no-limit Texas hold’em poker event. The second AI called Libratus was written in the United States by Carnegie Mellon University. Libratus won against all four of his opponents. Here are a few official news outlet reporting on the subject:
- MIT Technology Review on DeepStack
- MIT Technology Review on AI playing Poker
- Carnegie Mellon University on Libratus
- The Verge on Libratus
Crash Course: UI Design Here is one useful quote from the very visually appealing post: ” UX design is what makes an interface useful, and UI design is what makes an interface beautiful“.
What Are Some Dirty Tactics from Academia? It was a surprising, but educational read. A lot of dirty tricks when it comes to publishing research papers. I am sure most people are not like this, but we can also confirm that not everyone has the advancement of science as priority.