Although AI offers vast implications for engineering, production, supply chain, customer experience, and mobility services, progress in AI-driven transformation has been sluggish and uneven due to lingering roadblocks. The number of automotive companies deploying AI at scale has grown from 7% in 2017 to 10% today, with OEMs generally making better progress than suppliers or dealers. Geographically, the US, where 25% of companies implement AI at scale, is leading the way in terms of progress, followed by the UK (14%) and Germany (12%). In terms of pronounced growth, China is making huge strides, having nearly doubled its share of scaled AI implementations, from 5% to 9%. The new report by the Capgemini Research Institute, Accelerating automotive's AI transformation: how driving AI enterprise-wide can turbo-charge organizational value, surveyed 500 executives from large automotive organizations in eight countries and interviewed a number of industry experts and entrepreneurs to understand how progress in deploying AI at scale can be accelerated.

The future of healthcare is bright. Daily articles and news reports herald the arrival of digital technologies and artificial intelligence right into the heart of our homes and healthcare institutions. Indeed, technology is not so much an add-on but seen to be essential to the NHS as we look to future proof our precious health service. The recent Topol Review explored how to prepare the workforce for a digital future, and the release of the "Code of Conduct for Artificial Intelligence Systems used by the NHS" amongst others are both statements of intent in order to deliver this ambitious digital agenda. In addition, Matt Hancock, secretary of state for health and social care, has also identified technology as the answer to some of the challenges faced by the NHS and this demonstrates the weight of commitment to artificial intelligence and digital health.

The Centers for Medicare and Medicaid Services has launched a new contest it hopes will speed the development of new artificial intelligence technologies that can better predict health outcomes and boost quality of care. WHY IT MATTERS CMS says the Artificial Intelligence Health Outcomes Challenge – announced by the agency on Wednesday, in partnership with American Academy of Family Physicians and the Laura and John Arnold Foundation – seeks to uncover and "unleash" new and innovative tools to help with the push toward value-based care. To do that, CMS is calling on developers from all industries to create new predictive AI applications to help providers participating in CMS Innovation Center models to deliver better care and make quality measures more impactful. "The Artificial Intelligence Health Outcomes Challenge is a three stage competition that will begin with the Launch Stage, in which participants will submit an application at ai.cms.gov," "Up to 20 participants will be selected to participate in Stage 1 of the Challenge. We anticipate that more information about Stage 1 and Stage 2 will be announced later this year."

Prevencio Inc. announced data confirming the high accuracy of its artificial intelligence (AI)-driven, multiple-protein HART CVE Test for predicting cardiovascular events (CVE) and HART CAD Test for diagnosing coronary artery disease (CAD). Researchers believe these findings, presented at the 2019 American College of Cardiology (ACC) Scientific Sessions, March 16-18 in New Orleans, demonstrate the robustness and accuracy of these tests. The new data, from two additional hospitals, confirm results previously published from Massachusetts General Hospital and James Januzzi, M.D. HART CVE data was presented as "Validation of a Novel AI-driven Multi-biomarker Panel for Accurate Prediction of Incident Cardiovascular Events in Patients with Suspected Myocardial Infarction." The study was conducted by the University Heart Center in Hamburg, Germany and led by Dirk Westermann, M.D., Ph.D., head of the Structural Heart Program. In this study, 748 patients presenting to the emergency room suspected of having a heart attack were followed for one year to assess future cardiac events, including heart attack and cardiac death.

Robots with artificial intelligence (AI) might have once seemed like an idea only possible in science fiction, but as technology advances, chances are they will become more common. One such robot, for example, spoke to members of the United Kingdom Parliament last year about caring for the elderly. More recently, researchers at the University of Nottingham in the UK developed what's known as a "machine learning algorithm" -- think about it as a robot brain -- capable of learning from reams of data and then making predictions based on the data. In a new study, the algorithm was able to predict the risk of premature death in a group of middle aged people with better accuracy than a human with statistical models. The AI also took a fraction of the usual time to do so.

As much as we'd like to, nobody knows exactly what the future holds. However, several individuals have put their minds, time, and effort toward figuring out what the future of marketplaces, industry, and work will look like to the greatest possible degree of accuracy. For those who are looking to figure out which industries are most ripe for disruption, which are dinosaurs, and how the future of work applies in your own life, these forward-looking influencers are must-follows. But don't just stop at their Twitter pages, as these influencers have personal websites, TED Talks, and more that are well worth checking out. This influencer has racked up several job titles on his way to becoming one of the most popular voicers in the future of work sphere.

To learn who's taking home the Turing Award, people might turn to their trusted talking bots, like Siri or Alexa. Or, in fact, some of the very technology the three winners helped bring to life. Yoshua Bengio, Geoffrey Hinton and Yann LeCun have earned what's often referred to as the Nobel Prize of the tech world for their pioneering work in artificial intelligence, the Association for Computing Machinery announced Wednesday. The researchers, working both independently and together, helped advance the thinking and application of neural networks, the technology that gives computers the ability to recognize patterns, interpret language and glean insights from complex data. "Artificial intelligence is now one of the fastest-growing areas in all of science and one of the most talked-about topics in society," Cherri Pancake, president of the computing society, said in a statement.

Meta-learning methods leverage past experience to learn data-driven inductive biases from related problems, increasing learning efficiency on new tasks. This ability renders them particularly suitable for sequential decision making with limited experience. Within this problem family, we argue for the use of such approaches in the study of model-based approaches to Bayesian Optimisation, contextual bandits and Reinforcement Learning. We approach the problem by learning distributions over functions using Neural Processes (NPs), a recently introduced probabilistic meta-learning method. This allows the treatment of model uncertainty to tackle the exploration/exploitation dilemma. We show that NPs are suitable for sequential decision making on a diverse set of domains, including adversarial task search, recommender systems and model-based reinforcement learning.

Running hydrodynamical simulations to produce mock data of large-scale structure and baryonic probes, such as the thermal Sunyaev-Zeldovich (tSZ) effect, at cosmological scales is computationally challenging. We propose to leverage the expressive power of deep generative models to find an effective description of the large-scale gas distribution and temperature. We train two deep generative models, a variational auto-encoder and a generative adversarial network, on pairs of matter density and pressure slices from the BAHAMAS hydrodynamical simulation. The trained models are able to successfully map matter density to the corresponding gas pressure. We then apply the trained models on 100 lines-of-sight from SLICS, a suite of N-body simulations optimised for weak lensing covariance estimation, to generate maps of the tSZ effect. The generated tSZ maps are found to be statistically consistent with those from BAHAMAS. We conclude by considering a specific observable, the angular cross-power spectrum between the weak lensing convergence and the tSZ effect and its variance, where we find excellent agreement between the predictions from BAHAMAS and SLICS, thus enabling the use of SLICS for tSZ covariance estimation.

Kriging is the predominant method used for spatial prediction, but relies on the assumption that predictions are linear combinations of the observations. Kriging often also relies on additional assumptions such as normality and stationarity. We propose a more flexible spatial prediction method based on the Nearest-Neighbor Neural Network (4N) process that embeds deep learning into a geostatistical model. We show that the 4N process is a valid stochastic process and propose a series of new ways to construct features to be used as inputs to the deep learning model based on neighboring information. Our model framework outperforms some existing state-of-art geostatistical modelling methods for simulated non-Gaussian data and is applied to a massive forestry dataset. GPs are used directly to model Gaussian data and as the basis of non-Gaussian models such as generalized linear (e.g., Diggle et al., 1998), quantile regression (e.g., Lum et al., 2012; Reich, 2012) and spatial extremes (e.g., Cooley et al., 2007; Sang and Gelfand, 2010) models. Similarly, Kriging is the standard method for geostatistical prediction.