In the age of digital transformation, global insurance provider AXA has adopted a decentralised approach to innovation. AXA International and New Markets (AXA INM) takes charge of AXA's operations in emerging and developing markets, covering Eastern European territories, Latin America, the GCC (Gulf Cooperation Council), Africa, India, Singapore, Malaysia and more. "There's a significant amount of Transformation to deliver across the 20-25 entities," says Kuldeep Kaushik, Chief Operating and Transformation Officer at AXA INM. "We have very different maturity levels across the businesses and very different technology landscapes as well. Part of my role is evaluating each of those entities and defining programmes which are specific to their maturity, business strategy, and needs and capabilities."

In this paper we address the problem of solving ill-posed inverse problems in imaging where the prior is a neural generative model. Specifically we consider the decoupled case where the prior is trained once and can be reused for many different log-concave degradation models without retraining. Whereas previous MAP-based approaches to this problem lead to highly non-convex optimization algorithms, our approach computes the joint (space-latent) MAP that naturally leads to alternate optimization algorithms and to the use of a stochastic encoder to accelerate computations. The resulting technique is called JPMAP because it performs Joint Posterior Maximization using an Autoencoding Prior. We show theoretical and experimental evidence that the proposed objective function is quite close to bi-convex. Indeed it satisfies a weak bi-convexity property which is sufficient to guarantee that our optimization scheme converges to a stationary point. Experimental results also show the higher quality of the solutions obtained by our JPMAP approach with respect to other non-convex MAP approaches which more often get stuck in spurious local optima.

Due to their accuracies, methods based on ensembles of regression trees are a popular approach for making predictions. Some common examples include Bayesian additive regression trees, boosting and random forests. This paper focuses on honest random forests, which add honesty to the original form of random forests and are proved to have better statistical properties. The main contribution is a new method that quantifies the uncertainties of the estimates and predictions produced by honest random forests. The proposed method is based on the generalized fiducial methodology, and provides a fiducial density function that measures how likely each single honest tree is the true model. With such a density function, estimates and predictions, as well as their confidence/prediction intervals, can be obtained. The promising empirical properties of the proposed method are demonstrated by numerical comparisons with several state-of-the-art methods, and by applications to a few real data sets. Lastly, the proposed method is theoretically backed up by a strong asymptotic guarantee.

Graph convolution is the core of most Graph Neural Networks (GNNs) and usually approximated by message passing between direct (one-hop) neighbors. In this work, we remove the restriction of using only the direct neighbors by introducing a powerful, yet spatially localized graph convolution: Graph diffusion convolution (GDC). GDC leverages generalized graph diffusion, examples of which are the heat kernel and personalized PageRank. It alleviates the problem of noisy and often arbitrarily defined edges in real graphs. We show that GDC is closely related to spectral-based models and thus combines the strengths of both spatial (message passing) and spectral methods. We demonstrate that replacing message passing with graph diffusion convolution consistently leads to significant performance improvements across a wide range of models on both supervised and unsupervised tasks and a variety of datasets. Furthermore, GDC is not limited to GNNs but can trivially be combined with any graph-based model or algorithm (e.g. spectral clustering) without requiring any changes to the latter or affecting its computational complexity. Our implementation is available online.

This blog post has been contributed to the Salesforce Blog by one of our Dreamforce '19 Innovator sponsors. Technology is at an inflection point. In this next chapter of digital transformation, companies must reinvent their entire business with data to create a more personal and relevant customer experience. Yet while today's companies are awash with data, unlocking its value requires them to change how they operate and make decisions. That's where artificial intelligence (AI) comes in.

Vital: The Future of Healthcare is an anthology of short stories. Vital has already gathered stories from leading futurist writers, weaving together disparate visions of what comes next in health and health science. Our visions of the future -- whether dark or hopeful, thrilling or mundane -- have always challenged us to examine our world. What challenges will we face? Vital: The Future of Healthcare aims to explore these questions as they relate to humanity's physical and mental well-being.

Newswire) VSBLTY Groupe Technologies Corp. (CSE: VSBY) (5VS.F) (VSBGF) ("VSBLTY"), a leading retail software technology company, and Energetika, an international provider of "intelligent lighting" solutions, have begun deployment of their smart city security contract which combines Energetika's smart lighting with VSBLTY's crowd analytics and facial recognition to help keep Mexico City's neighborhoods safe. Energetika CEO Rodrigo Calderon said, "We have begun phase one deployment of security kits covering up to 40,000 endpoints throughout 56 communities in Mexico City beginning in the boroughs of Miguel Hidalgo, Cuajimalpa, Benito Juarez and Cuauhtemoc. Each neighborhood security kit consists of high definition cameras equipped with VSBLTY facial recognition and analytics, wireless alarms, motion sensors and panic buttons integrated with high LED facade light fixtures. This low cost system runs off local citizens' internet service and is accessible on their mobile devices in real time. With this unique security kit deployment model perfected, we have introduced this program to other Central and South American municipalities whose needs are equally compelling and where this cost-efficient solution can be installed in three million security cameras or more."

"I say this to everyone in the media world who I talk to," says Darren Atkins, wrapping up our phone interview: "Please, absolutely do not portray this as a hidden agenda to get rid of staff." Atkins is the Chief Technology Office for AI automation at East Suffolk and North Essex NHS Foundation Trust – group of hospitals employing more than 10,000 staff, who serve a quarter of a million people in the South East of England. "If this technology is applied in the wrong way, it can be very threatening," Atkins says. "Our main priority is to free up time for staff to do the work that they should be doing, rather than the work that has no value." Just over a year ago, Atkins led the deployment of virtual workers across his group of NHS hospitals – and according to him, it's been an unqualified success. Patients are missing fewer appointments and staff are happier.

Between 2013 and 2016, U.S. farmers and ranchers weathered a 45% dip in net farm income -- the largest since the Great Depression -- while the number of mouths to feed grew sharply by the day. The global population is expected to increase by 2.2 billion by 2050, and the world's farmers will have to grow about 70% more food than is now produced. If you ask Microsoft, the solution lies in technology. The tech giant's FarmBeats program, which launched in preview today on Azure Marketplace ahead of Ignite 2019, is a multi-year effort to bring robust data analytics to the agriculture sector. With a backend built on Azure and compatibility with hardware from a range of top manufacturers, it aims to promote what Ranveer Chandra, FarmBeats project lead and chief scientist at Azure Global, calls "data-driven" farming techniques. The International Food Policy Research Institute claims these can boost farm productivity by as much as 67% while reducing resource usage.

"Banning killer robots is both politically savvy and morally necessary," said Mary Wareham, the Arms Division advocacy director at Human Rights Watch and coordinator of the Campaign to Stop Killer Robots. "European states should take the lead and open ban treaty negotiations if they are serious about protecting the world from this horrific development." Countries attending the annual meeting of states parties to the Convention on Conventional Weapons (CCW) at the United Nations in Geneva will decide on November 15 whether to continue diplomatic talks on killer robots, also known as lethal autonomous weapons systems or fully autonomous weapons. Since 2014, these states have held eight meetings on lethal autonomous weapons systems under the auspices of the Convention on Conventional Weapons (CCW), a major disarmament treaty. Over the course of those meetings, states have built a shared understanding of concern, but they have struggled to reach agreement on credible recommendations for multilateral action due to the objections of a handful of military powers, most notably Russia and the United States.