We consider high-dimensional regression over subgroups of observations. Our work is motivated by biomedical problems, where disease subtypes, for example, may differ with respect to underlying regression models, but sample sizes at the subgroup-level may be limited. We focus on the case in which subgroup-specific models may be expected to be similar but not necessarily identical. Our approach is to treat subgroups as related problem instances and jointly estimate subgroup-specific regression coefficients. This is done in a penalized framework, combining an $\ell_1$ term with an additional term that penalizes differences between subgroup-specific coefficients. This gives solutions that are globally sparse but that allow information-sharing between the subgroups. We present algorithms for estimation and empirical results on simulated data and using Alzheimer's disease, amyotrophic lateral sclerosis and cancer datasets. These examples demonstrate the gains our approach can offer in terms of prediction and the ability to estimate subgroup-specific sparsity patterns.

Mixture of Experts (MoE) is a popular framework in the fields of statistics and machine learning for modeling heterogeneity in data for regression, classification and clustering. MoE for continuous data are usually based on the normal distribution. However, it is known that for data with asymmetric behavior, heavy tails and atypical observations, the use of the normal distribution is unsuitable. We introduce a new robust non-normal mixture of experts modeling using the skew $t$ distribution. The proposed skew $t$ mixture of experts, named STMoE, handles these issues of the normal mixtures experts regarding possibly skewed, heavy-tailed and noisy data. We develop a dedicated expectation conditional maximization (ECM) algorithm to estimate the model parameters by monotonically maximizing the observed data log-likelihood. We describe how the presented model can be used in prediction and in model-based clustering of regression data. Numerical experiments carried out on simulated data show the effectiveness and the robustness of the proposed model in fitting non-linear regression functions as well as in model-based clustering. Then, the proposed model is applied to the real-world data of tone perception for musical data analysis, and the one of temperature anomalies for the analysis of climate change data. The obtained results confirm the usefulness of the model for practical data analysis applications.

Mixture of Experts (MoE) is a popular framework for modeling heterogeneity in data for regression, classification, and clustering. For regression and cluster analyses of continuous data, MoE usually use normal experts following the Gaussian distribution. However, for a set of data containing a group or groups of observations with heavy tails or atypical observations, the use of normal experts is unsuitable and can unduly affect the fit of the MoE model. We introduce a robust MoE modeling using the $t$ distribution. The proposed $t$ MoE (TMoE) deals with these issues regarding heavy-tailed and noisy data. We develop a dedicated expectation-maximization (EM) algorithm to estimate the parameters of the proposed model by monotonically maximizing the observed data log-likelihood. We describe how the presented model can be used in prediction and in model-based clustering of regression data. The proposed model is validated on numerical experiments carried out on simulated data, which show the effectiveness and the robustness of the proposed model in terms of modeling non-linear regression functions as well as in model-based clustering. Then, it is applied to the real-world data of tone perception for musical data analysis, and the one of temperature anomalies for the analysis of climate change data. The obtained results show the usefulness of the TMoE model for practical applications.

Chances are you either love or hate your car -- and soon the feeling could be mutual. Japanese automaker Honda will showcase a concept car at the Consumer Electronics Show next month that is capable of understanding the driver's emotions and developing emotions of its own, the company announced this week. The company provided few details as to how the technology will work or alter the driving experience. But we do know that the concept car, called the NeuV, is being touted as an automated electric vehicle that includes an "emotion engine." That's the name for artificial intelligence that Honda says will "enable machines to artificially generate their own emotions."

Astro Teller has an unusual way of starting a new project: He tries to kill it. Teller is the head of X, formerly Google X, the advanced technology lab of Alphabet. At X's headquarters not far from the Googleplex in Mountain View, Calif., Teller leads a group of engineers, inventors, and designers devoted to futuristic "moonshot" projects like self-driving cars, delivery drones, and Internet-beaming balloons. To turn their wild ideas into reality, Teller and his team have developed a unique approach. It starts with trying to prove that whatever it is that you're trying to do can't be done--in other words, trying to kill your own idea. As Teller explains, "Instead of saying, 'What's most fun to do about this or what's easiest to do first?' we say, 'What is the most likely reason this project won't make it?' The ideas that survive get additional rounds of scrutiny, and only a tiny fraction eventually becomes official projects; the proposals that are found to have an Achilles' heel are ...

Despite the best efforts of movies like Ex Machina, Morgan and Avengers: Age of Ultron, a new survey found that people from around the world largely see artificial intelligence having a more positive than negative impact on their lives and society in general. Communications firm Weber Shandwick has just published its "AI-Ready or Not: Artificial Intelligence Here We Come!" report, conducted with KRC Research, for marketers, surveying 2,100 consumers across five global markets on AI, its many uses, how they see it evolving, and how comfortable they are with that development. But don't go tearing up your plans for an unconnected cabin in the woods just yet, because even though consumer survey respondents were seven times more likely to see the sunny side of AI, a full one-third of respondents also admitted to knowing nothing about AI at all. The survey also interviewed 150 marketing executives (primarily CMOs) in the U.S., the U.K., and China responsible for the oversight and execution of marketing or branding activities at their organizations. On the consumer side, 77% of respondents would like AI's development to accelerate or remain at its current pace, two-thirds or more trust AI with handling medication reminders, travel directions, entertainment, targeted news, and manual labor and mechanics.

SAN FRANCISCO – Forget the Terminator. The next robot on the horizon may be wearing a lab coat. Artificial intelligence (AI) is already helping scientists form testable hypotheses that enable experts to run real experiments, and the technology may soon be poised to help businesses make decisions, one scientist says. However, that doesn't mean the machines will be taking over from humans entirely. Instead, humans and machines have complementary skillsets, so AI could help researchers with the work they already do, Laura Haas, a computer scientist and director of the IBM Research Accelerated Discovery Lab in San Jose, California, said here Wednesday (Dec.

IBM provided an Analytics Assessment & Insights Impact Grant to the Shenzhen CDC to help further their mission of infectious disease prevention and control, shouldering the monitoring, alarming and treatment of emergency public health events in Shenzhen for the 12 million citizenry by building a self adaptive online machine learning module that provides cognitive-based modeling for epidemic disease prediction and analysis on case number and trend. The output of this work has allowed the organization to develop prediction models to help forecast seasonal flu outbreaks and provide information to citizens on affected areas. CDC organizations across China are now considering the implementation of this solution to help track flu and other infectious diseases.

US consumers appear to be warming up to the idea of using of chatbots as a customer relationship management (CRM) tool, according to new research from Aspect Software Research. In its online survey of more than 1,000 18- to 65-year-old US consumers, 44% said that if a company could get the experience right, they would prefer to use a chatbot or automated experience for CRM. That's up four percentage points from the share of respondents who noted the same response in 2015. Chatbots can best be thought of as software programs that use messaging as the interface to carry out various tasks for users. They're generally integrated into messaging apps to capitalize on these apps' vast reach and the conversational interaction they promote.

Delivering objects via drone is a tempting notion bound by hard constraints: drones are small, so the cargo has to be small. Drones need power to fly, and any additional weight requires more power to cover the same distance, which further limits the size of the cargo. For a drone delivery to make sense, then, the small cargo has to justify both its weight and the urgency of a drone flight. Pound for pound and ounce for ounce, few cargoes match that limitation better than blood. In a study published in the journal Transfusion, Johns Hopkins researcher Timothy Amukele demonstrated that drones are a safe and efficient way to get blood pouches to remote locations.