Interesting articles from the worlds of personalization, machine learning, and data science that caught my eye this past week.
Personalized music streaming service Pandora is leveraging its personalization tech to serve up more relevant ads to its listeners.
“It’s becoming quite apparent to us that the world of playing the perfect music to people and the world of playing perfect advertising to them are strikingly similar”. Link.
A very exciting technique called Deep Learning has come into the limelight over the past year. Teams wielding this technique have won first place in machine-learning/AI competitions, achieving breakthroughs in accuracy. The article below has a nice overview (HT to my colleague Yevgeniy Popkov for the reference).
“Using massive amounts of data to recognize photos and speech, deep-learning computers are taking a big step towards true artificial intelligence.” Link.
Google is at the forefront of using Deep Learning for lots of interesting things, such as automatically identifying house numbers from Street View images.
“The task of searching these images manually to spot and identify the numbers is not one anybody could approach with relish. So, naturally, Google has solved the problem by automating it … Their method turns out to rely on a neural network that contains 11 levels of neurons that they have trained to spot numbers in images.” Link.
If Deep Learning piques your interest, check this out. If you are in the mood for something technical, read this by one of the pioneers of the field. CQuotient has R&D in progress applying Deep Learning to personalization and we will blog about this work later in the year.