You're whizzing along a dark, outback highway when the symptoms start. Your mouth is so dry it's getting hard to swallow or talk, your grip on the steering wheel is weakening and that centre white line just turned into two. You pull in at a small town hospital where the tired junior doctor tells you it's probably a virus, and then hands your case over to night staff. She retakes your history wearing Google Glass, and the web-enabled headgear uses voice recognition to input your symptoms into a massive data base. It's botulism; very rare, often fatal, and you're just in time to get the antitoxin.

One crucial goal in kernel online learning is to bound the model size. Common approaches employ budget maintenance procedures to restrict the model sizes using removal, projection, or merging strategies. Although projection and merging, in the literature, are known to be the most effective strategies, they demand extensive computation whilst removal strategy fails to retain information of the removed vectors. An alternative way to address the model size problem is to apply random features to approximate the kernel function. This allows the model to be maintained directly in the random feature space, hence effectively resolve the curse of kernelization. However, this approach still suffers from a serious shortcoming as it needs to use a high dimensional random feature space to achieve a sufficiently accurate kernel approximation. Consequently, it leads to a significant increase in the computational cost. To address all of these aforementioned challenges, we present in this paper the Dual Space Gradient Descent (DualSGD), a novel framework that utilizes random features as an auxiliary space to maintain information from data points removed during budget maintenance. Consequently, our approach permits the budget to be maintained in a simple, direct and elegant way while simultaneously mitigating the impact of the dimensionality issue on learning performance. We further provide convergence analysis and extensively conduct experiments on five real-world datasets to demonstrate the predictive performance and scalability of our proposed method in comparison with the state-of-the-art baselines.

While neural machine translation (NMT) is making good progress in the past two years, tens of millions of bilingual sentence pairs are needed for its training. However, human labeling is very costly. To tackle this training data bottleneck, we develop a dual-learning mechanism, which can enable an NMT system to automatically learn from unlabeled data through a dual-learning game. This mechanism is inspired by the following observation: any machine translation task has a dual task, e.g., English-to-French translation (primal) versus French-to-English translation (dual); the primal and dual tasks can form a closed loop, and generate informative feedback signals to train the translation models, even if without the involvement of a human labeler. In the dual-learning mechanism, we use one agent to represent the model for the primal task and the other agent to represent the model for the dual task, then ask them to teach each other through a reinforcement learning process. Based on the feedback signals generated during this process (e.g., the language-model likelihood of the output of a model, and the reconstruction error of the original sentence after the primal and dual translations), we can iteratively update the two models until convergence (e.g., using the policy gradient methods). We call the corresponding approach to neural machine translation \emph{dual-NMT}. Experiments show that dual-NMT works very well on English$\leftrightarrow$French translation; especially, by learning from monolingual data (with 10\% bilingual data for warm start), it achieves a comparable accuracy to NMT trained from the full bilingual data for the French-to-English translation task.

Apple has stayed true to its promise and published its first academic paper on artificial intelligence. The world's most valuable company has traditionally kept its AI research private but earlier this month Ruslan Salakhutdinov, director of AI research at Apple, made a pledge to start being more open. The new Apple paper -- published December 22 and titled "Learning from simulated and unsupervised images through adversarial training" -- gives an insight into some of the techniques that Apple is using to develop AI. In the study, which was published through the Cornell University Library, Apple researchers explain a technique that can be used to improve how an algorithm learns to "see" what is in an image. The paper's six authors state that using synthetic images (such as those seen in a video game), as opposed to real-world images, can be more efficient when it comes to training AI models known as neural networks, which are designed to think in the same way as the human brain. Because synthetic image data is already labelled and annotated while real-world images aren't.

In 2016, the White House recognized the importance of AI at its Frontiers Conference. The concept of driverless cars became a reality, with Uber's self-driving fleet in Pittsburgh and Tesla's new models equipped with the hardware for full autonomy. Google's DeepMind platform, AlphaGo, beat the world champion of the game--10 years ahead of predictions. "Increasing use of machine learning and knowledge-based modeling methods" are major trends to watch in 2017, said Marie desJardins, associate dean and professor of computer science at the University of Maryland, Baltimore County. How will this play out?

As the end of the year looms over the horizon, it's time to take a look forward and mull over the next big thing in technology. Like any year before, 2017 will bring its own problems and solutions, shaping up both the way we use and think about technology. So without further ado, take a dive into the future and check out some of the most exciting tech trends to look forward to in 2017. With forecasts predicting its growth into a $30 billion market as early as 2020, much has been said about the bright future of virtual reality. Although the technology remained on the verge of mainstream culture throughout most of 2015, things finally started to pick up over the last 12 months – and it seems this time around VR might legitimately reach the masses next year.

In the retail insurance space, consumers are more connected than ever, via a multitude of devices and through multiple platforms. This has two consequences: first, consumers increasingly expect a much better, smarter service. "People are frustrated with a clunky process for buying insurance, and want an easier and quicker process through simple digital channels," says John Salmon, a technology partner at Hogan Lovells. Second, a larger proportion of consumers fall into Generation Y or the millennial generation: these individuals are less likely to own property or cars, are less attracted by life assurance and are looking for more tailored cover they can buy easily and quickly. They are attracted by the sharing economy.

On Tuesday, the US government continued a national conversation on the impact of artificial intelligence on the workforce. The new White House report, Artificial Intelligence, Automation, and the Economy, serves as a follow-up to its October report, Preparing for the Future of Artificial Intelligence, which looked at the role of government in the development and implementation of artificial intelligence. It's an effort by the US government to address the huge impact automation is currently having on jobs--which promises to be felt more deeply as artificial intelligence advances. The report explores the history and impact of automation on the economy, and looks at jobs that could be lost or gained from artificial intelligence. It also outlines three policy strategies meant to help prevent automation from taking jobs away from humans.

A Queensland startup has celebrated a new deal with a major European customer after it perfected new deep learning technology that recognises images beyond simple shape and texture matching. TrademarkVision's deal with the European Union Intellectual Property Office this month comes after a successful beta test that saw around 1,000 trademark image searches conducted each day. The EUIPO becomes the first governmental agency to take up the technology, although the terms of the contract were not disclosed. In the past year, the company has been working on deep learning technology to take its software to the next level of intelligence, and the move has borne fruit in a spectacular way with the EU deal. "We've focused on machine learning techniques so the system can recognise objects in trademarks and logos much like humans do. Despite the wide variety of ways humans pictorially depict objects in logos, deep learning has helped to solve this semantic challenge in a quick and robust way," said TrademarkVision founder and chief executive Sandra Mau.

Each of these trucks is the size of a small two-story house. None has a driver or anyone else on board. Mining company Rio Tinto has 73 of these titans hauling iron ore 24 hours a day at four mines in Australia's Mars-red northwest corner. At this one, known as West Angelas, the vehicles work alongside robotic rock drilling rigs. The company is also upgrading the locomotives that haul ore hundreds of miles to port--the upgrades will allow the trains to drive themselves, and be loaded and unloaded automatically.