— In the News —
Peter Norvig, Google's Research Director, talks about his long experience in Artificial Intelligence and Machine Learning at the recent Vienna Gödel Lecture. This is a very clear, amusing, and engaging presentation (starts at 10:56). Highly recommended.
Y Combinator is one of the most important accelerators for technology startups. Since 2005, it has funded over 800 new businesses with a combined valuation of over 30 billion dollars. Here's a peek into the latest big data and analytics startups from last week's Demo Day.
— Tools and Techniques —
Spark is currently the most popular and fastest growing Apache project. It's advertised to be lightning fast and because it avoids disk I/O and can keep everything in memory, is especially well suited to iterative algorithms like those used in machine learning. This is a good overview of what you can do with Spark and it's core libraries.
Great article that explores the massive landscape of tools that are available to data scientists. At a high-level, these are organized by where they are in the workflow: getting data, wrangling data and analyzing data.
Books are great for learning theory but a lot can go wrong in practice. Here are the three key things that the books won't tell you.
— Resources —
Curated collection of machine learning frameworks, libraries and software. This collection is maintained by Joseph Misiti on Github where it's been forked 900 times and has over 5000 stars! This looks pretty complete but if you have a favorite that isn't on the list already, let him know.
Google's Deep Mind is working at the cutting edge of general‑purpose learning algorithms. This collection of publications describes what they're doing there.
— Data Viz —
D3 is one of the most important libraries for enabling interactive data visualizations on the web. This is a well-organized and comprehensive collection of resources for learning and mastering D3. To get an idea of what D3 is capable of, check out Mike Bostock's Visualizing Algorithms in this week's Archive Picks (below). Even if you're not interested in building D3 applications, it's becoming ubiquitous for web-based data viz work and is worth being familiar with.
— Archive Pick —
This article by Mike Bostock is a masterpiece. Don't miss it.
— Career —
If you're either looking for or thinking about looking for a data science job, this is a MUST READ.
And if you're on the other side of the hiring table, you'll DEFINITELY want to check this article out too.