— In the News —
At a recent conference in Puerto Rico, an unprecedented meeting of the minds discussed ethics in Artificial Intelligence and signed an open letter pledging to conduct AI research that is robust and beneficial. Attendees included industry heavyweights such as Elon Musk, Oxford AI ethicist Nick Bostrom, Skype founder Jaan Tallinn and Google AI expert Shane Legg. This is a great overview of why they're concerned and how deep those concerns really go.
Google is into a lot of things you may not expect. It has recently been helping an organization called Imazon to create monthly deforestation maps of the Amazon. Many more such projects are in the works, tackling issues as varied as overfishing, sea-ice melt, and predicting the spread of disease. Here's a good overview of what's to come and why everyone is not a fan of this work.
— Tools and Techniques —
Great Deep Learning Tutorial from Stanford, including exercises and code.
The Twitter engineering team recently released an open-source R package called AnomalyDetection to automatically detect big data anomalies in a practical and robust way. This is a great overview of why it's important and how it works.
— Resources —
Here's a curated collection of free books, lectures, websites, datasets, papers, and frameworks for Deep Learning enthusiasts. This is updated regularly and a not-to-miss resource.
Data Innovation Day is an online conference of data scientists, technologists, civic leaders, and other experts exploring how data-driven innovation is helping to grow the economy and create a better world. There's a wide variety of topics including open data, new business models, data viz, civic hacking, and wearables. January 22 from 12:00 PM – 4:00 PM EST. Looks very worthwhile and it's free.
Free online data journalism course comprised of video lectures, tutorials, assignments, readings and discussion forums. Video lectures by industry leaders anchor each of five modules. This had been expected to close last month but was recently extended due to high demand. Now available through March 31.
— Data Viz —
Interesting immersive data experiment using virtual reality goggles. Even if you don't have VR goggles, this is worth checking out since this is surely a preview of things to come for data visualization.
Nice overview of arc diagrams, including examples and things to watch out for.