Artificial intelligence (AI) has continued to gain prominence in 2017 as one of the biggest upcoming technologies. It is beginning to have more of an influence on companies' strategies and is predicted to drive significant change for organisations. Every conference this year contains a dead human genius reincarnated as software system or a robot. Yes, there is a lot of hype, but there is real worth in AI and Machine Learning. Read our counseling on how to avoid adopting "black box" approach.

Plus, the new season of Black Mirror is here. Out-of-warranty battery replacements will now cost only $29.Apple apologizes for confusion over slowdowns with older iPhones Apple has been in hot water for the last few weeks after the company admitted that it sometimes reduced processor speeds on iPhones with aging batteries as a way to balance performance and battery life. Today, the company is apologizing for not being more transparent with its customers and released more details on how exactly iOS manages battery and performance. That, however, hasn't stopped several lawsuits, from the US to France. And what we're most excited about in 2018.The best games of 2017 Early 2017 brought us legitimate contenders for game of the year in The Legend of Zelda: Breath of the Wild, Horizon Zero Dawn and Persona 5 -- and that's not to mention Resident Evil's return to form.

Artificial intelligence (AI) was one of 2017's hottest industry buzzwords as many have begun turning to machines to solve problems that are simply too large for humans to calculate. Once upon a time, AI was an academic pursuit -- but now it has become more affordable and attainable to pursue on a smaller scale, opening it up to use by a variety of companies for a variety of purposes. Feedzai recently told PYMNTS that Big Data paved the way for this shift, and that by 2020, U.S. companies could be saving as much as $60 billion thanks to the help of AI and machine learning. Business management consultancy Accenture expects AI to add $8.3 trillion in economic activity for the U.S. by 2035. It's clear that this trend is building some significant momentum in the payments space and adjacent industries.

To the editor: News about the environment has been so sad lately. On the front page of Friday's Los Angeles Times, you had a story about park rangers being killed by elephant poachers in Congo, and on the Opinion page you had an article about Congress opening up the Arctic National Wildlife Refuge in Alaska to oil drilling. We have already lost a large portion of our natural resources, but it seems that humans will not be satisfied until all of them are gone. This is not just an American problem, it is a worldwide problem. Only a small percentage of the global population is working on conservation, and we are losing.

Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the environment as they solve simple tasks. That is, we've used these techniques to build closed-loop systems rather than open-loop ones as before. The simulator need not match the real-world in appearance or dynamics; instead, we randomize relevant aspects of the environment, from friction to action delays to sensor noise. Our new results provide more evidence that general-purpose robots can be built by training entirely in simulation, followed by a small amount of self-calibration in the real world. This robot was trained in simulation with dynamics randomization to push a puck to a goal.

A novel data-driven stochastic robust optimization (DDSRO) framework is proposed for optimization under uncertainty leveraging labeled multi-class uncertainty data. Uncertainty data in large datasets are often collected from various conditions, which are encoded by class labels. Machine learning methods including Dirichlet process mixture model and maximum likelihood estimation are employed for uncertainty modeling. A DDSRO framework is further proposed based on the data-driven uncertainty model through a bi-level optimization structure. The outer optimization problem follows a two-stage stochastic programming approach to optimize the expected objective across different data classes; adaptive robust optimization is nested as the inner problem to ensure the robustness of the solution while maintaining computational tractability. A decomposition-based algorithm is further developed to solve the resulting multi-level optimization problem efficiently. Case studies on process network design and planning are presented to demonstrate the applicability of the proposed framework and algorithm.

Apple and Amazon are in licensing discussions with Riyadh on investing in Saudi Arabia, sources claim. The move is part of Crown Prince Mohammed bin Salman's push to give the conservative kingdom a high-tech look. A third source confirmed to Reuters that Apple was in talks with SAGIA, Saudi Arabia's foreign investment authority. Both companies already sell products in Saudi Arabia via third parties but they and other global tech giants have yet to establish a direct presence. Apple and Amazon are in licensing discussions with Riyadh on investing in Saudi Arabia, sources claim.

Microsoft's Cognitive Toolkit (better known as CNTK) is a commercial-grade and open-source framework for deep learning tasks. At present CNTK does not have a native R interface but can be accessed through Keras, a high-level API which wraps various deep learning backends including CNTK, TensorFlow, and Theano, for the convenience of modularizing deep neural network construction. The latest version of CNTK (2.1) supports Keras. The RStudio team has developed an R interface for Keras making it possible to run different deep learning backends, including CNTK, from within an R session. This tutorial illustrates how to simply and quickly spin up a Ubuntu-based Azure Data Science Virtual Machine (DSVM) and to configure a Keras and CNTK environment.

About 20 years later, a young Mary Shelley answered a dare to write a ghost story, which she shared at a small gathering at Lake Geneva. Her story would go on to be published as a novel, "Frankenstein; or, the Modern Prometheus," on Jan. 1, 1818. Both are stories about our powers to create things that take on a life of their own. Goethe's poem comes to a climax when the apprentice calls out in a panic: While the master fortunately returns just in time to cancel the treacherous spell, Shelley's tale doesn't end so nicely: Victor Frankenstein's monster goes on a murderous rampage, and his creator is unable to put a stop to the carnage. That's the question we face on the 200th anniversary of "Frankenstein," as we find ourselves grappling with the unintended consequences of our creations on Facebook, to artificial intelligence and human genetic engineering.

This paper considers inference over distributed linear Gaussian models using factor graphs and Gaussian belief propagation (BP). The distributed inference algorithm involves only local computation of the information matrix and of the mean vector, and message passing between neighbors. Under broad conditions, it is shown that the message information matrix converges to a unique positive definite limit matrix for arbitrary positive semidefinite initialization, and it approaches an arbitrarily small neighborhood of this limit matrix at a doubly exponential rate. A necessary and sufficient convergence condition for the belief mean vector to converge to the optimal centralized estimator is provided under the assumption that the message information matrix is initialized as a positive semidefinite matrix. Further, it is shown that Gaussian BP always converges when the underlying factor graph is given by the union of a forest and a single loop. The proposed convergence condition in the setup of distributed linear Gaussian models is shown to be strictly weaker than other existing convergence conditions and requirements, including the Gaussian Markov random field based walk-summability condition, and applicable to a large class of scenarios.