This particular class of algorithms has recently gained popularity due to their effectiveness in solving large-scale optimization problems in machine learning, compressed sensing, image processing, and computational statistics. Coordinate descent algorithms solve optimization problems by successively minimizing along each coordinate or coordinate hyperplane, which is ideal for parallelized and distributed computing. Avoiding detailed technicalities and proofs, this monograph gives relevant theory and examples for practitioners to effectively apply coordinate descent to modern problems in data science and engineering. To keep the primer up-to-date, we intend to publish this monograph only after no additional topics need to be added and we foresee no further major advances in the area. 1 Introduction

Deep learning methods such as multitask neural networks have recently been applied to ligand-based virtual screening and other drug discovery applications. Using a set of industrial ADMET datasets, we compare neural networks to standard baseline models and analyze multitask learning effects with both random cross-validation and a more relevant temporal validation scheme. We confirm that multitask learning can provide modest benefits over single-task models and show that smaller datasets tend to benefit more than larger datasets from multitask learning. Additionally, we find that adding massive amounts of side information is not guaranteed to improve performance relative to simpler multitask learning. Our results emphasize that multitask effects are highly dataset-dependent, suggesting the use of dataset-specific models to maximize overall performance.

The focus in this paper is Bayesian system identification based on noisy incomplete modal data where we can impose spatially-sparse stiffness changes when updating a structural model. To this end, based on a similar hierarchical sparse Bayesian learning model from our previous work, we propose two Gibbs sampling algorithms. The algorithms differ in their strategies to deal with the posterior uncertainty of the equation-error precision parameter, but both sample from the conditional posterior probability density functions (PDFs) for the structural stiffness parameters and system modal parameters. The effective dimension for the Gibbs sampling is low because iterative sampling is done from only three conditional posterior PDFs that correspond to three parameter groups, along with sampling of the equation-error precision parameter from another conditional posterior PDF in one of the algorithms where it is not integrated out as a "nuisance" parameter. A nice feature from a computational perspective is that it is not necessary to solve a nonlinear eigenvalue problem of a structural model. The effectiveness and robustness of the proposed algorithms are illustrated by applying them to the IASE-ASCE Phase II simulated and experimental benchmark studies. The goal is to use incomplete modal data identified before and after possible damage to detect and assess spatially-sparse stiffness reductions induced by any damage. Our past and current focus on meeting challenges arising from Bayesian inference of structural stiffness serve to strengthen the capability of vibration-based structural system identification but our methods also have much broader applicability for inverse problems in science and technology where system matrices are to be inferred from noisy partial information about their eigenquantities.

Inverse classification is the process of perturbing an instance in a meaningful way such that it is more likely to conform to a specific class. Historical methods that address such a problem are often framed to leverage only a single classifier, or specific set of classifiers. These works are often accompanied by naive assumptions. In this work we propose generalized inverse classification (GIC), which avoids restricting the classification model that can be used. We incorporate this formulation into a refined framework in which GIC takes place. Under this framework, GIC operates on features that are immediately actionable. Each change incurs an individual cost, either linear or non-linear. Such changes are subjected to occur within a specified level of cumulative change (budget). Furthermore, our framework incorporates the estimation of features that change as a consequence of direct actions taken (indirectly changeable features). To solve such a problem, we propose three real-valued heuristic-based methods and two sensitivity analysis-based comparison methods, each of which is evaluated on two freely available real-world datasets. Our results demonstrate the validity and benefits of our formulation, framework, and methods.

A child's first words are celebrated, but once we are lingual, we just open our mouths and speak. But how did we come to have smooth, spoken language? Scientists have turned to our relatives, non-human primates, for answers. In the hunt for what makes us unique, researchers have some ideas. One distinction they have long highlighted is the position of humans' voice box. Humans have a low larynx, while other primates have a high larynx.

With the development of free, open-source machine learning and artificial intelligence tools like Google's TensorFlow and sci-kit learn, as well as "ML-as-a-service" products like Google's Cloud Prediction API and Microsoft's Azure Machine Learning platform, it's never been easier for companies of all sizes to harness the power of data. But machine learning is such a vast, complex field. Where do you start learning how to use it in your business? In this article, we'll survey the current landscape of machine learning algorithms and explain how they work, provide example applications, share how other companies use them, and provide further resources on learning about them. This executive overview will provide the first step in learning how to apply machine learning algorithm(s) to make your business more efficient, more effective, and more profitable.

Entertainment consulting firm Accenture has been doing its digital consumers survey for about a decade, but its 2017 edition was the first to include questions around artificial intelligence and technology. And the results were surprising to the company. "There's pretty intense interest in AI this year, and a year ago there wasn't anyone talking about it," said Charles Hartley, Accenture's global media and analyst relations manager for communications, media and high-tech businesses. "Consumers aren't intimidated by it at all, apparently. The survey--which interviewed just under 26,000 consumers across 26 countries--found that 62% of people are comfortable with AI apps--like Amazon's Alexa--responding to a voice query, even though only 4% of people actually own a standalone, digital, voice-enabled device (like the Amazon Echo or Google Home) as of the end of 2016. Nearly 90% of respondents said that artificial intelligence simply makes it easier to do things, and a third said they're interested in using voice-enabled digital assistants available in smartphones. Hartley made special note of the 68% who deemed AI "less biased" than humans and the 64% who said AI "communicates more politely." More than half said AI is "less likely to make a mistake." Meanwhile, on Jan. 11, LinkedIn founder Reid Hoffman, the journalism-centric Knight Foundation and others announced they've created a $27 million fund to research AI applications for the public, with MIT's Media Lab and the Berkman Klein Center for Internet & Society at Harvard University serving as academic research stations. The hope is to bring in a wide range of tech and academic voices to the future of AI applications, the groups said in a statement. "Artificial intelligence agents will impact our lives in every society on Earth.

CB Insights revealed the AI 100 winners during The Innovation Summit in Santa Barbara, a gathering of top executives and investors to explore the industries of the future. The CB Insights research team selected the winners based on a combination of data submitted by the companies, responses to interview questions and the company's Mosaic Score. Mosaic is an algorithm built with funding from the National Science Foundation that gives predictive intelligence into the health of private companies. "At Ayasdi, we see clearly that intelligent applications enabled by AI and big data will be as transformative for knowledge-based processes as the industrial revolution was for mechanical processes," said Ayasdi CEO Gurjeet Singh. "Intelligent applications running on our platform are already becoming a fundamental part of our customer's operations with breakthrough results. We are delighted to be recognized by CB Insights for the success our clients are having in deploying Ayasdi to use AI to solve real-world business problems at scale."

I had a fun ride attending a very interesting lecture this semester called Programming Paradigms. I learned about the four main paradigms that exist: imperative, object-oriented, functional and logic programming. Now, I'm sure every developer has heard about imperative, OO and functional, but to be honest I had no idea what logic programming was about. I was intrigued, what could this paradigm I had never heard about be, what does it excel in and could it be useful for day-to-day programming problems? The book The Pragmatic Programmer has a tip called "Invest Regularly in Your Knowledge Portfolio": Different languages solve the same problems in different ways.

Do you think Hollywood has lost touch with American values? Do you think Hollywood has lost touch with American values? The contentious presidential campaign was filled with accusations of elitism and bias by the media -- from the news to entertainment. Many supporters of Donald J. Trump saw his victory as a repudiation of the so-called liberal elite. So as 2017 begins, we ask: Is Hollywood representing all Americans? Are Hollywood values out of sync with American values? It's the start of a conversation we'll have all year with Hollywood's creators, consumers and observers. Most of all, we want to hear from you. Is Hollywood out of touch with your America? Here's what our critics and writers have to say: KENNETH TURAN on potent Hollywood visions that helped elect Trump TV's affluent bubble: MARY McNAMARA on Hollywood's reluctance to deal with class issues Fear of the powerful woman: JUSTIN CHANG on working women and men still behaving badly Realistic or cliche?: JEFFREY FLEISHMAN on ...