There are data scientists that are doing truly new stuff. I've been in the analytics for over 20 years (PhD in stats, 1993), and what I do now - including the methodology used, not just the data - is radically different from what I did even 5 years ago. It involves a lot of automation, new algorithms (Jackknife regression, model-free confidence intervals, hidden decision trees, brand new random number generation, feature selection, predictive power) applied to all sorts of data, usually big data sets. It actually goes far beyond processing data: data processing is the tip of the iceberg, but the big picture of what I'm doing is making real-time systems (such as traffic generatiion for this very website) work automatically and smoothly, and in a scalable way. From scratch and/or using vendor platforms - I design, build, and deploy systems that work, with home-made metrics used to test performance and find areas of improvement.

Health Catalyst has created Healthcare.ai, a website that offers free open source predictive analytics software for hospitals and other healthcare organizations. "Wherever you have a data set that you pull together, you can create a model based on that by using these tools," said Levi Thatcher, director of data science at Health Catalyst. Machine learning and predictive analytics to improve health outcomes has so far been limited to an elite group of data scientists, mostly in the nation's top academic medical centers, he pointed out. Healthcare.ai – open source predictive analytics software – is part of a mission to make machine learning accessible to the thousands of healthcare professionals with only basic technical skills, but who share an interest in using the technology to improve patient care, Thatcher explained. By making its central repository of proven machine learning algorithms freely available, Healthcare.ai opens the doors to a large, diverse group of technical healthcare professionals to quickly use machine learning tools to build accurate models.

How can we programme a robot to behave morally when we don't have a working definition for morality ourselves? This is one of the many questions involved with the field of artificial intelligence and the development of advanced robots. While it might sound like something out of a science fiction novel, artificial intelligence has quickly become a facet of our daily lives. Take Siri or Google Now on our smartphones, and Amazon's Alexa. Rudimentary as they are today, these so-called "smart assistants" represent the future of artificial intelligence.

In statistical modeling, there are various algorithms to build a classifier, and each algorithm makes a different set of assumptions about the data. For Big Data, it pays off to analyze the data upfront and then design the modeling pipeline accordingly. Statistical modeling is a lot like engineering. In engineering, there are various ways to build a key-value storage, and each design makes a different set of assumptions about the usage pattern. In statistical modeling, there are various algorithms to build a classifier, and each algorithm makes a different set of assumptions about the data.

Yiling Chen (yiling@seas.harvard.edu) is Gordon McKay Professor of Computer Science at Harvard University, Cambridge, MA. Arpita Ghosh (arpitaghosh@cornell.edu) is an associate professor of information science at Cornell University, Ithaca, NY. Michael Kearns (mkearns@cis.upenn.edu) is a professor and National Center Chair of Computer and Information Science at the University of Pennsylvania, Philadelphia, PA. Tim Roughgarden (tim@cs.stanford.edu) is an associate professor of CS at Stanford University, Stanford, CA. Jennifer Wortman Vaughan (jenn@microsoft.com) is a senior researcher at Microsoft Research, New York, NY.

I was eager to learn about the latest developments in the Semantic Web through the lens of a "new kind of semantics" as Abraham Bernstein et al. explored in their Viewpoint "A New Look at the Semantic Web" (Sept. If I understand it correctly, semantics is a mapping function that leads from manifest expressions to elements in a given arbitrary domain. Based on set theory, logicians have developed a framework to set up such mapping for formal languages like mathematics, provided one can fix an interpretation function. On the other hand, 20th-century logicians (notably Alfred Tarski) warned of the limits of the framework when applied to human languages. Now, to the extent it embraces a set-theoretic semantics (as in the W3C's Ontology Web Language), the Semantic Web seems to be facing exactly such limitations or experiencing, dealing with, and suffering them.

Changing geopolitical strategy and rapid technological progress are also conspiring to render the term military increasingly incoherent. The United States--the one power to emerge unscathed from World War II and ascendant from the Cold War, and the one power that invests the greatest amount of resources in its battle-hardened military--possesses such overwhelming conventional military superiority over anyone else that potential adversaries instead embrace asymmetric warfare. Thus, Gen. Valery Gerasimov, chief of the General Staff of the Armed Forces of Russia, wrote in a seminal article in 2014 that "the very'rules of war' have changed. The role of nonmilitary means of achieving political and strategic goals has ... in many cases ... exceeded the power of force of weapons in their effectiveness." He also spoke of the need for "the broad use of political, economic, informational, humanitarian, and other nonmilitary measures," with conventional force "resorted to ... primarily for the achievement of final success in the conflict."

This month, I will attend the Conference and Workshop on Neural Information Processing Systems (NIPS), the premier gathering in the machine learning field. I've participated in this conference most years since it began in 1987 and I'm looking forward once again to catching up with colleagues and friends as well as exploring new developments in the field. Until recently, the conference attracted a few hundred attendees. The number of participants has grown rapidly in recent years and this year there are more than 4,500 people registered! This explosion of activity in machine learning is remarkable and reflects the positive trend of research making its way to the marketplace.

The U.S. Internet giant announced that Google Translate has been switched to a new system called Neural Machine Translation (NMT), an end-to-end learning framework that learns from millions of examples. The multilingual system is based on machine learning that provides computers with the ability to learn without being explicitly programmed. Unlike the current translation system that was adopted 10 years ago, the new system considers the entire sentence as one unit. Previous systems translated words and phrases independently within a sentence. Google said the NMT interprets entire sentences, making the translation not only sound much more like a native speaker of the language but more accurate.

Amazon Web Services CEO Andy Jassy introduced a bevy of new services and capabilities at the AWS re:Invent conference in Las Vegas this week. The new analytic and artificial intelligence (AI) services aren't unique, but there's little doubt they'll be huge hits. Jassy framed his announcements around the theme of giving enterprises "superpowers." Examples included powerful new compute instances supporting superhero-like speed, new database services enabling "flight" from the high cost of commercial databases, and new IoT services enabling "shapeshifting" out to the edge of the enterprise. I was most interested in the "X-Ray Vision" introductions, which included Athena and QuickSight analytic services and Rekognition, Polly and Lex artificial intelligence (AI) services.