Mark Zuckerberg has defended the rights of holocaust deniers to stay on Facebook – because they are being genuine. The Facebook boss said that he found the belief that the holocaust did not happen was deeply offensive. But he said that the people using his site to promote should be allowed to use it and that the posts should stay up. There are many things that people get wrong and those that claim that the holocaust did not happen are one of them, he suggested during an interview. He claimed that since the people are mistaken in their belief, rather than intending to harm anyone, they will continue to be allowed to post on the site.

A couple of years ago, as Brian Brackeen was preparing to pitch his facial recognition software to a potential customer as a convenient, secure alternative to passwords, the software stopped working. Panicked, he tried adjusting the room's lighting, then the Wi-Fi connection, before he realized the problem was his face. Brackeen is black, but like most facial recognition developers, he'd trained his algorithms with a set of mostly white faces. He got a white, blond colleague to pose for the demo, and they closed the deal. It was a Pyrrhic victory, he says: "It was like having your own child not recognize you."

Huawei Consumer Business Group (CBG), the global smartphone giant, has announced the launch of its new HUAWEI nova 3 and HUAWEI nova 3i in the UAE. Both smartphones are powered by Artificial Intelligence (AI) in its overall functioning and camera features. With dual front camera (24MP 2MP), users can expect the best-in-technology'selfie' that a smartphone can offer. HUAWEI nova 3 series was initially designed with the younger and trendier consumers in mind, for whom a smartphone is not just part of their lifestyle but also represents their personal style. The series is designed to offer an AI-enhanced lifestyle, which includes a huge focus on an outstanding selfie experience with its AI beautification features and front camera that allows one to capture AI selfies.

This story was co-published with NPR. But dig deeper and the implications of what they're selling might give many patients pause: A future in which everything you do -- the things you buy, the food you eat, the time you spend watching TV -- may help determine how much you pay for health insurance. With little public scrutiny, the health insurance industry has joined forces with data brokers to vacuum up personal details about hundreds of millions of Americans, including, odds are, many readers of this story. The companies are tracking your race, education level, TV habits, marital status, net worth. Then they feed this information into complicated computer algorithms that spit out predictions about how much your health care could cost them. Are you a woman who recently changed your name? You could be newly married and have a pricey pregnancy pending. Or maybe you're stressed and anxious from a recent divorce.

Hundreds of organisations and thousands of techies, including Elon Musk, Demis Hassabis from Google's DeepMind, and the head of the Chocolate Factory's AI lab Jeff Dean have promised never to support the development of autonomous weapons. The pledge was organised by the Future of Life Institute, an outreach geroup focused on tackling existential risks. It was co-founded by a group of researchers, including Max Tegmark, a physics professor at the Massachusetts Institute of Technology, Viktoriya Krakovna, a scientist at DeepMind, and Jann Tallinn, co-founder of Skype. "We will neither participate in nor support the development, manufacture, trade, or use of lethal autonomous weapons," it reads. The promise is based on a "moral component" that machines should be forbidden from making "life-taking decisions."

Some would say it's the magic of a festival - stumbling upon a random stage and accidentally discovering your new favourite band. You could call it following your festival instinct. But what if you ditched all that and did the complete opposite? What if you took arguably the most nerdy thing in the world - statistics - and used it to try to have the best festival experience ever? I consulted a stats expert, packed up a giant bar chart, and headed to 2000 Trees in Gloucestershire to find out. And - just a warning - this article is incredibly, incredibly geeky.

On paper Israel and China are unlikely close trading partners. China, the world's second-largest country, is the biggest exporter on the planet. While Israel, a tiny strip of land in the Middle East, is only in 45th place on the global exporting league table. And importantly - Israel has always been a steadfast ally of the US. So given the current trading spat between the US and China, you would expect Israel to be firmly on the American side.

Ultrasound (US) is the most widely used fetal imaging technique. However, US images have limited capture range, and suffer from view dependent artefacts such as acoustic shadows. Compounding of overlapping 3D US acquisitions into a high-resolution volume can extend the field of view and remove image artefacts, which is useful for retrospective analysis including population based studies. However, such volume reconstructions require information about relative transformations between probe positions from which the individual volumes were acquired. In prenatal US scans, the fetus can move independently from the mother, making external trackers such as electromagnetic or optical tracking unable to track the motion between probe position and the moving fetus. We provide a novel methodology for image-based tracking and volume reconstruction by combining recent advances in deep learning and simultaneous localisation and mapping (SLAM). Tracking semantics are established through the use of a Residual 3D U-Net and the output is fed to the SLAM algorithm. As a proof of concept, experiments are conducted on US volumes taken from a whole body fetal phantom, and from the heads of real fetuses. For the fetal head segmentation, we also introduce a novel weak annotation approach to minimise the required manual effort for ground truth annotation. We evaluate our method qualitatively, and quantitatively with respect to tissue discrimination accuracy and tracking robustness.

Systematic compositionality is the ability to recombine meaningful units with regular and predictable outcomes, and it's seen as key to humans' capacity for generalization in language. Recent work has studied systematic compositionality in modern seq2seq models using generalization to novel navigation instructions in a grounded environment as a probing tool, requiring models to quickly bootstrap the meaning of new words. We extend this framework here to settings where the model needs only to recombine well-trained functional words (such as "around" and "right") in novel contexts. Our findings confirm and strengthen the earlier ones: seq2seq models can be impressively good at generalizing to novel combinations of previously-seen input, but only when they receive extensive training on the specific pattern to be generalized (e.g., generalizing from many examples of "X around right" to "jump around right"), while failing when generalization requires novel application of compositional rules (e.g., inferring the meaning of "around right" from those of "right" and "around").

Generative Adversarial Networks (GANs) can produce images of surprising complexity and realism, but are generally modeled to sample from a single latent source ignoring the explicit spatial interaction between multiple entities that could be present in a scene. Capturing such complex interactions between different objects in the world, including their relative scaling, spatial layout, occlusion, or viewpoint transformation is a challenging problem. In this work, we propose to model object composition in a GAN framework as a self-consistent composition-decomposition network. Our model is conditioned on the object images from their marginal distributions to generate a realistic image from their joint distribution by explicitly learning the possible interactions. We evaluate our model through qualitative experiments and user evaluations in both the scenarios when either paired or unpaired examples for the individual object images and the joint scenes are given during training. Our results reveal that the learned model captures potential interactions between the two object domains given as input to output new instances of composed scene at test time in a reasonable fashion.