— In the News —
THIS IS AWESOME. Prediction algorithms and some engineering magic guarantees a bullseye every time.
There's a lot of talk these days about how artificial intelligence will revolutionize medicine. In some fields, there are already reports of A.I.s being as good or better than human experts at some types of diagnoses. But on a practical level, what's really possible? This longread in the New Yorker explores the opportunities and the realities of what's been accomplished so far.
— Sponsored Link —
Find Top Developers: We help companies like Airbnb, Pfizer, and Artsy find great developers. Let us find your next great hire. Get started today.
— Tools and Techniques —
Entity resolution is the task of disambiguating records across and within datasets. The applications are huge but as you'll see in this article, so are the challenges. This is a thorough introduction to entity resolution: what it is, why it's important, and how to resolve conflicts in your data using the Python Dedupe package.
Great introduction to stock market analysis using R. This is the first of a two-part series that covers how to obtain stock data using the quantmod package, visualizing stock data, moving averages, developing a moving-average crossover strategy, backtesting, and benchmarking.
If you ever need to make maps, Natural Earth is a key resource. Datasets are public domain, are available at multiple scales in both vector and raster format, and they include GIS attributes. The rnaturalearth package makes it easy to access and work with Natural Earth map data using R.
— Deep Learning —
Here's an easy way to get all those open source libraries installed and working on your own computer. This virtual machine image is complete with Ubuntu, Python 3.5, all the required libraries, and tools like TensorFlow, Theano, Keras, OpenCV, and dlib.
Ruslan Salakhutdinov, who leads Apple’s AI efforts, says emerging techniques could make the most popular approach in the field far more powerful.
Instacart is using Emoji math to save hundreds of years of shopping time. No joke. Here's how they're doing it.
This encrypted deep learning scheme protects privacy, prevents AI theft, and has some interesting implications for superintelligence.