File photo - A Google carpet is seen at the entrance of the new headquarters of Google France before its official inauguration in Paris, France Dec. 6, 2011. A recently surfaced Google video discusses a creepy concept for collecting vast quantities of user data that could span generations. The video, which was obtained by The Verge, paints an unsettling picture of how data could theoretically be harnessed on an epic scale. The Verge reports that the video was produced in 2016 by X, a research and development subsidiary of Google's parent company Alphabet. X, formerly known as Google X, describes itself as a "moonshot factory" focused on developing technologies to make the world "a radically better place."

Banks Must Invest in Reskilling Their Workforces to Seize AI-driven Growth Opportunities, Accenture Report Finds Stronger commitment to AI could boost revenues 34 percent and employment 14 percent by 2022 NEW YORK; May 2, 2018 – Although bank leaders recognize that intelligent technologies are reshaping the core banking process and can transform customer experiences, few plan to significantly increase investments in reskilling their workforces to enable these technologies in the near-term, according to a report by Accenture (NYSE: ACN). Based on two surveys – one of 100 banking executives and another of 1,300 non-executive bank employees – the report, "Future Workforce Survey - Banking: Realizing the Full Value of AI," estimates that if banks invest in artificial intelligence (AI) and human-machine collaboration at the same rate as top-performing companies do, they could boost revenues by 34 percent and raise employment levels 14 percent by 2022. "As AI becomes more nuanced, its role in banks is moving beyond automation to elevating human capabilities," said Alan McIntyre, a senior managing director at Accenture and head of the company's Banking practice. "To benefit from the potential of AI, banks need to implement'applied intelligence' – combining technology and human ingenuity – across all areas of their core business. To achieve this, they will need commitment from the highest levels of leadership and an understanding that this evolution will require a dramatic change in their workforce."

There's often a fine line between hero and villain, and by most accounts, artificial intelligence (AI) is on the villain side, sucking jobs out of the economy. These days you can't throw a rock without hitting some pundit prognosticating on the millions of jobs that will be lost from AI. One oft-cited Oxford University study predicted 47% of jobs are in jeopardy. But while AI conjures up robots and dystopian science fiction movies, it isn't magic. Today's AI consists of algorithms developed with "training data" that improve over time, otherwise known as machine learning.

Don't even worry about Dyret the robot. At first glance, the scrawny quadruped looks pathetic, as it struggles to walk without collapsing. But keep watching, and you'll see it start to improve--walking slowly, yet ever more proficiently. Dyret the robot is teaching itself to walk. Machines like Cassie the biped or SpotMini the robot dog are quickly mastering locomotion, thanks to line after line of meticulous code.

UAVs are tackling everything from disease control to vacuuming up ocean waste to delivering pizza, and more. Drone technology has been used by defense organizations and tech-savvy consumers for quite some time. However, the benefits of this technology extends well beyond just these sectors. With the rising accessibility of drones, many of the most dangerous and high-paying jobs within the commercial sector are ripe for displacement by drone technology. The use cases for safe, cost-effective solutions range from data collection to delivery. And as autonomy and collision-avoidance technologies improve, so too will drones' ability to perform increasingly complex tasks. According to forecasts, the emerging global market for business services using drones is valued at over $127B. As more companies look to capitalize on these commercial opportunities, investment into the drone space continues to grow. A drone or a UAV (unmanned aerial vehicle) typically refers to a pilotless aircraft that operates through a combination of technologies, including computer vision, artificial intelligence, object avoidance tech, and others. But drones can also be ground or sea vehicles that operate autonomously.

Without constant learning and adapting, businesses risk becoming dinosaurs. The world, after all, is in constant flux, driven by the technology needed to help integrate into organisations. According to a recent (December 2017) report from the McKinsey Global Institute ('Jobs Lost Jobs Gained'), "Automation will bring big shifts to the world of work as AI and robotics change or replace some jobs but others are created". In the UK, 20% of current work activities will be automated by 2030. Some authors claim that as little as 35% of current skills will still be relevant in five years – others put it at even fewer.

Scientists from Imperial College London and the University of Edinburgh have developed software capable of detecting dementia and stroke precursors using CT scans. The development, according to the study's lead author and clinical lecturer at Imperial College London, Dr Paul Bentley, "could lead to better treatments and care for patients in everyday practice". The research team's machine learning software has been trained using 1,082 CT scans of stroke patients from 70 hospitals across the UK between 2000 and 2014. The software is able to identify and measure a marker of Small Vessel Disease (SVD), a common precursor to strokes and dementia, which reduces blood flow to the brain's deep white matter connections. Doctors currently rely on CT and/or MRI scans to detect SVD.

Here's a question: do you consider yourself to be a trusting person? Or let me put it another way: would you put your life in the hands of a total stranger? This morning I woke up. I switched on the light – trusting that I wouldn't be electrocuted by a faulty lamp, or cord, or socket. I prepared my breakfast – trusting that I wouldn't be poisoned by salmonella in my factory-processed muesli.

We apply neural nets with ReLU gates in online reinforcement learning. Our goal is to train these networks in an incremental manner, without the computationally expensive experience replay. By studying how individual neural nodes behave in online training, we recognize that the global nature of ReLU gates can cause undesirable learning interference in each node's learning behavior. We propose reducing such interferences with two efficient input transformation methods that are geometric in nature and match well the geometric property of ReLU gates. The first one is tile coding, a classic binary encoding scheme originally designed for local generalization based on the topological structure of the input space. The second one (EmECS) is a new method we introduce; it is based on geometric properties of convex sets and topological embedding of the input space into the boundary of a convex set. We discuss the behavior of the network when it operates on the transformed inputs. We also compare it experimentally with some neural nets that do not use the same input transformations, and with the classic algorithm of tile coding plus a linear function approximator, and on several online reinforcement learning tasks, we show that the neural net with tile coding or EmECS can achieve not only faster learning but also more accurate approximations. Our results strongly suggest that geometric input transformation of this type can be effective for interference reduction and takes us a step closer to fully incremental reinforcement learning with neural nets.

We introduce a distributionally robust maximum likelihood estimation model with a Wasserstein ambiguity set to infer the inverse covariance matrix of a $p$-dimensional Gaussian random vector from $n$ independent samples. The proposed model minimizes the worst case (maximum) of Stein's loss across all normal reference distributions within a prescribed Wasserstein distance from the normal distribution characterized by the sample mean and the sample covariance matrix. We prove that this estimation problem is equivalent to a semidefinite program that is tractable in theory but beyond the reach of general purpose solvers for practically relevant problem dimensions $p$. In the absence of any prior structural information, the estimation problem has an analytical solution that is naturally interpreted as a nonlinear shrinkage estimator. Besides being invertible and well-conditioned even for $p>n$, the new shrinkage estimator is rotation-equivariant and preserves the order of the eigenvalues of the sample covariance matrix. These desirable properties are not imposed ad hoc but emerge naturally from the underlying distributionally robust optimization model. Finally, we develop a sequential quadratic approximation algorithm for efficiently solving the general estimation problem subject to conditional independence constraints typically encountered in Gaussian graphical models.