IN JULY 2011 Sebastian Thrun, who among other things is a professor at Stanford, posted a short video on YouTube, announcing that he and a colleague, Peter Norvig, were making their "Introduction to Artificial Intelligence" course available free online. By the time the course began in October, 160,000 people in 190 countries had signed up for it. At the same time Andrew Ng, also a Stanford professor, made one of his courses, on machine learning, available free online, for which 100,000 people enrolled. Both courses ran for ten weeks. Such online courses, with short video lectures, discussion boards for students and systems to grade their coursework automatically, became known as Massive Open Online Courses (MOOCs).

Knoema is the latest data provider to add a conversational interface. The McLean, Virginia-based company has launched an artificial intelligence (AI)-powered chatbot called Yodatai, which it describes as the first digital assistant for public and corporate data. The name, CEO Vladimir Bougay told me, is a shortening of "your data AI." His company provides access to industry, governmental and market data from thousands of providers, including the US Census, the US Department of Energy and other sources around the world. Additionally, Yodatai has been integrated with product analytics platform Amplitude and can access other databases via API.

Modeling causal dependencies often demands cycles at a coarse-grained temporal scale. If Bayesian networks are to be used for modeling uncertainties, cycles are eliminated with dynamic Bayesian networks, spreading indirect dependencies over time and enforcing an infinitesimal resolution of time. Without a ``causal design,'' i.e., without anticipating indirect influences appropriately in time, we argue that such networks return spurious results. By identifying activator random variables, we propose activator dynamic Bayesian networks (ADBNs) which are able to rapidly adapt to contexts under a causal use of time, anticipating indirect influences on a solid mathematical basis using familiar Bayesian network semantics. ADBNs are well-defined dynamic probabilistic graphical models allowing one to model cyclic dependencies from local and causal perspectives while preserving a classical, familiar calculus and classically known algorithms, without introducing any overhead in modeling or inference.

In the wake of the recent Manchester terrorist attacks, in which 22 people--mostly parents, teenagers, and children as young as eight years old--were murdered by a suicide bomber, the question that lingered on so many people's minds is, "how?" How could he, the bomber, do it? It's a question that many people have found themselves asking too often, not just after highly publicized attacks in the U.S., but also in countries like Kenya and Nigeria where terror attacks by militant groups such as Boko Haram and al-Shabaab attract less global attention but are no less deadly, or heart-wrenchingly awful. And yet, despite behaviors that many would label immoral, terrorists often couch their activities in moral terms--invoking concepts such as "social cleansing" and "moral purification," attacking people and symbols that they believe are representative of moral failings. But how can people allegedly motivated by morality engage in behaviors that, from the outside, appear to be so clearly immoral? Agustín Ibáñez, a cognitive science researcher at Argentina's Ineco Foundation at Favaloro University, and Adolfo Garcia, a researcher at the National Scientific and Technical Research Council (CONICET), took an interesting tact towards understanding how the mind of a terrorist differs from the mind of, well, people who don't commit acts of terror.

New system for i-vector speaker recognition based on variational autoencoder (VAE) is investigated. VAE is a promising approach for developing accurate deep nonlinear generative models of complex data. Experiments show that VAE provides speaker embedding and can be effectively trained in an unsupervised manner. LLR estimate for VAE is developed. Experiments on NIST SRE 2010 data demonstrate its correctness. Additionally, we show that the performance of VAE-based system in the i-vectors space is close to that of the diagonal PLDA. Several interesting results are also observed in the experiments with $\beta$-VAE. In particular, we found that for $\beta\ll 1$, VAE can be trained to capture the features of complex input data distributions in an effective way, which is hard to obtain in the standard VAE ($\beta=1$).

But modern tools are helping to revive Ireland's national language. An Irish proverb advises that it is often wise for one to hold his tongue. An té is ciúine is é is buaine, or "he who is silent is the stronger." But that ancestral wisdom isn't the best policy when the very language it comes from is threatened. The Irish language, Gaelic, is one of more than 40 percent of the world's 6,000 spoken languages that are endangered, according to UNESCO.

Opinions expressed by Forbes Contributors are their own. The author is a Forbes contributor. The opinions expressed are those of the writer. Unless you've been on a sabbatical deep in the rainforests of Peru, you've probably heard about Artificial Intelligence (AI). But if you still relate it to all things science fiction and robotic, it's time to look further.

It is a truth universally acknowledged that a single chatbot, powered by artificial intelligence, will inevitably learn from human users to be a sexist pig. Microsoft learned this lesson the hard way in 2016 when it released the Twitter chatbot known as Tay. Within hours, Tay was tweeting sentences like "Zoe Quinn is a Stupid Whore" and "I fucking hate feminists and they should all die and burn in hell." A recent study published in the journal Science, conducted by researchers at Princeton University and the University of Bath, found machine learning inherently "absorbs stereotyped biases" from human internet users. That includes racial bias, as proved by Google search algorithms' propensity to prioritize images of white women and babies over people of color.

This is a sample of the data science projects I have been working on my own. The Zika Project, is an extensive analysis of microcephaly cases related to Zika in Brazil. This case study tries to explain how weather conditions from January 2015 to May 2016, projected 2015 and 2016 total population of men and women within a reproductive age (15- 44), prevalence of microcephaly cases, growth rate of microcephaly, and sanitation and demographic characteristics of the 27 Brazilian states have influenced the increase of microcephaly confirmed reported cases linked to zika from February 2016 to May 2016. To describe and report variables/features with greater emphasis on microcephaly, the study uses linear regression, lasso and ridge regression, regression trees, random forest regression and gradient boosting regressor. This is analysis unveils what factors other than elevation and runners split's strategy are better predictors of finishing within the top 15 male and female runners of the 2016 Boston Marathon In this short analysis explains, I used a expanded version of the mincer equation and find that marital status, gender, student's province of residence and country where student pursued his/her postgraduate studies are complementary features to explain the return of income/investement.

The smokestack of a refinery stands next to the Mantaro River in La Oroya, Peru. One of the things that has intrigued me the most about deep learning image cataloging algorithms is their ability to watch the world go by at scale each day through the incredible volume of news and social media images that are generated from every corner of the world and essentially generate a live ground truthed catalog of what's happening moment by moment. Of particular interest for disaster response and environmental monitoring is the ability of such algorithms to recognize imagery of flooding, drought, smog, litter, destruction, violence and other indicators of ongoing ground and air pollution and sudden natural disasters. What might a system look like? Two years ago I met Kadi Kenk, Head of Partnerships for "Let's do it" which is a social good organization founded in Estonia in 2008 that bills itself as a "social movement against trash."