We develop Riemannian Stein Variational Gradient Descent (RSVGD), a Bayesian inference method that generalizes Stein Variational Gradient Descent (SVGD) to Riemann manifold. The benefits are two-folds: (i) for inference tasks in Euclidean spaces, RSVGD has the advantage over SVGD of utilizing information geometry, and (ii) for inference tasks on Riemann manifolds, RSVGD brings the unique advantages of SVGD to the Riemannian world. To appropriately transfer to Riemann manifolds, we conceive novel and non-trivial techniques for RSVGD, which are required by the intrinsically different characteristics of general Riemann manifolds from Euclidean spaces. We also discover Riemannian Stein's Identity and Riemannian Kernelized Stein Discrepancy. Experimental results show the advantages over SVGD of exploring distribution geometry and the advantages of particle-efficiency, iteration-effectiveness and approximation flexibility over other inference methods on Riemann manifolds.

Learning from many real-world datasets is limited by a problem called the class imbalance problem. A dataset is imbalanced when one class (the majority class) has significantly more samples than the other class (the minority class). Such datasets cause typical machine learning algorithms to perform poorly on the classification task. To overcome this issue, this paper proposes a new approach Neighbors Progressive Competition (NPC) for classification of imbalanced datasets. Whilst the proposed algorithm is inspired by weighted k-Nearest Neighbor (k-NN) algorithms, it has major differences from them. Unlike k- NN, NPC does not limit its decision criteria to a preset number of nearest neighbors. In contrast, NPC considers progressively more neighbors of the query sample in its decision making until the sum of grades for one class is much higher than the other classes. Furthermore, NPC uses a novel method for grading the training samples to compensate for the imbalance issue. The grades are calculated using both local and global information. In brief, the contribution of this paper is an entirely new classifier for handling the imbalance issue effectively without any manually-set parameters or any need for expert knowledge. Experimental results compare the proposed approach with five representative algorithms applied to fifteen imbalanced datasets and illustrate this algorithms effectiveness.

Self-riding motorcycles and artificial intelligence are the future of riding, according to two Yamaha concepts. The Motobot Ver.2 and MOTOROiD will be among six debut models and concepts to be unveiled at the Tokyo Motor Show from October 27 to November 5. Yamaha unveiled its first Motobot in 2015 with the promise that it would be able to ride faster than MotoGP legend Valentino Rossi. We're not sure what this second version will be able to do or what its purpose would be! Yamaha will also unveil its wild-looking MOTOROiD with artificial intelligence that will allow the machine to identify its owner and interact with them. The experimental machine is aimed at personal mobility functions, the company says.

A new short film from the Campaign Against Killer Robots warns of a future where weaponised flying drones target and assassinate certain members of the public, using facial recognition technology to identify them. Is this a realistic threat that could rightly spur an effective ban on the technology? Or is it an overblown portrayal designed to scare governments into taking simplistic, unnecessary and ultimately futile action? We asked two academics for their expert opinions. A new short film from the Campaign Against Killer Robots warns of a future where weaponised flying drones target and assassinate certain members of the public, using facial recognition technology to identify them.

Robotics tries very hard to match the agility, versatility, and efficiency of animals. Some robots get very close in a few specific ways, but we're still chasing the dream of robots that can match our biological friends. One way of getting around this problem is by leveraging biology in the design of robots (and we do see a lot of bioinspiration in a variety of applications), but a more direct approach is to just make the robots themselves mostly biological. We've reported on this in the past in the context of flying insects, but this new cyborg beetle from Nanyang Technological University in Singapore is the smallest (and most controllable) yet. Here's how the Singaporean researchers, led by Professor Hirotaka Sato, describe their work in a recent paper: It is possible to use a living insect as a platform to develop a living insect-machine hybrid robot.

Since humanity is apparently gung-ho about bringing forth its own destruction, we are making rapid advancements in robotics technology. Just last month, a robot that once claimed it wanted to destroy humans was made a citizen of Saudi Arabia. We are living in a science fiction novel. On a more sincere note, this latest development from Massachusetts Institute of Technology and Harvard scientists is incredibly cool. Reported by the Los Angeles Times, they have engineered a type of artificial muscle that can lift up to 1,000 times its own weight.

Many self-driving car tests are conveniently run in warm, sunny climates where the road conditions are rarely less than ideal. But what about that significant chunk of the planet that gets snowfall? The Russian internet giant has started testing its autonomous Prius cars in winter conditions around Moscow's suburbs to see how they fare when snow obscures the roads and ice makes traction difficult. The video you see here is highly edited, but it suggests that the driverless machines are up to the job -- they can stay in their lanes, come to smooth stops and brake for pedestrians. These aren't the worst conditions a self-driving car could face.

A 95-year-old woman dubbed Western Australia's "real life tomb raider" and "Indiana Joan" in a profile earlier this month is now under investigation for how she acquired a trove of ancient artifacts. The woman, Joan Howard, was married to a senior diplomat with the United Nations in the Middle East. In the 1960s and 70s, she volunteered with American and British archaeologists, according to the West Australian. Given her diplomatic status, she traveled to several countries including Syria, Egypt, Lebanon, Jordan, Palestine and Israel looking for ancient artifacts, which she brought home to Perth, Australia. "It was all good fun. But as it turned out, very, very rewarding," Howard told the West Australian.

As smartphones move on to the next level of innovation ruled by Artificial Intelligence (AI) and Internet of Things (IoT) technologies, Samsung is fast developing future-ready devices to enhance consumer experience in India. The focus, says Asim Warsi, Head of Samsung India's Mobile Division who has now been promoted as the Global Vice President, is to build on services in the field of entertainment, utility and mobile payments for the end users. AI and IoT are very interesting spaces. There are a lot of possibilities and Samsung is in the thick of developing solutions, with giving a great hardware play on AI. We are building such platforms to enhance consumer experience and you will see more of them in 2018.

Baidu, China's answer to Google, and Xiaomi, which has modeled itself on the success of the America's most successful technology manufacturer in Apple, have announced a strategic partnership to tackle two emerging technologies, internet of things and artificial intelligence. Announced today at Xiaomi's first developer conference in Beijing, the alliance is a marriage between the third most valuable publicly traded Chinese technology company in Baidu and one of the world's most valuable private firms, Xiaomi. Clocking in with a market capitalization of $86 billion, Baidu is no slouch on the global technology stage -- with AI and deep learning among its core focuses -- while Xiaomi has become a key player in China and made inroads in emerging markets like India thanks to its affordable phones and smart devices which punch well above their price tag. The duo are announcing a collaboration for the first time, but TechCrunch understands that Baidu's conversational operating system -- DuerOS -- is already present in some Xiaomi hardware. That's little known because it was left unbranded, but now the companies have made their partnership official.