Last year China unveiled their ambitious plan to become the world leader in Artificial Intelligence (AI) research and deployment by 2030. In the roadmap the country's Central Government made it clear that not only did they want to expand their research and development capabilities but also find new ways to implement its use across every sector of their society, from industry, where it's being used to monitor the brain waves and emotional wellbeing of workers, to policing and defence, and even urban planning. Recently though it's been revealed that education is also a target and that several new AI innovations are already being tested in Chinese schools to "redefine how children are educated." While the broader deployment of AI and especially facial recognition software in many other countries is all too often mired in controversy, especially over concerns of accuracy, bias, and obviously personal privacy, the Chinese central government, which doesn't have to be beholden to anyone, has no such constraints. Now a report from People.cn, a state-run media news outlet, has revealed that a high school in Eastern China is testing a new facial recognition system that monitor students, and their levels of engagement, in real time.

HIROSHIMA – With technology powered by artificial intelligence, high school students are colorizing black and white photos of Hiroshima taken before the atomic bombing of the city in 1945. The 14 students at Hiroshima Jogakuin high school launched the initiative last November, aiming to make the images more vibrant and to revive the memories of survivors so they can better pass on their experiences to the next generation. "We are the last generation who can talk to atomic bomb survivors. We want to treasure conversations generated through the photos and contribute to keeping records of their accounts," said Anju Niwata, 16. The students began the work after learning about AI-based free colorization technology from their collaborator, Hidenori Watanabe, a professor of information design at the graduate school of the University of Tokyo.

One does not simply walk into TensorFlow. A PhD is a good opportunity for introspection. In fact, it is important to create opportunities for introspection no matter how busy or insignificant the present feels like. We should not regard our past as an immature period, but as an unfolding story. A story of discoveries, mistakes, skills, and projects that are now part of our professional consciousness.

Training deep neural networks requires many training samples, but in practice training labels are expensive to obtain and may be of varying quality, as some may be from trusted expert labelers while others might be from heuristics or other sources of weak supervision such as crowd-sourcing. This creates a fundamental quality-versusquantity tradeoff in the learning process. Do we learn from the small amount of high-quality data or the potentially large amount of weakly-labeled data? We argue that if the learner could somehow know and take the label-quality into account when learning the data representation, we could get the best of both worlds. To this end, we introduce "fidelity-weighted learning" (FWL) [9], a semi-supervised student-teacher approach for training deep neural networks using weakly-labeled data. FWL modulates the parameter updates to a student network (trained on the task we care about) on a per-sample basis according to the posterior confidence of its label-quality estimated by a teacher (who has access to the high-quality labels). Both student and teacher are learned from the data. We evaluate FWL on document ranking where we outperform state-of-the-art alternative semi-supervised methods.

A cognitive model of human learning provides information about skills a learner must acquire to perform accurately in a task domain. Cognitive models of learning are not only of scientific interest, but are also valuable in adaptive online tutoring systems. A more accurate model yields more effective tutoring through better instructional decisions. Prior methods of automated cognitive model discovery have typically focused on well-structured domains, relied on student performance data or involved substantial human knowledge engineering. In this paper, we propose Cognitive Representation Learner (CogRL), a novel framework to learn accurate cognitive models in ill-structured domains with no data and little to no human knowledge engineering. Our contribution is two-fold: firstly, we show that representations learnt using CogRL can be used for accurate automatic cognitive model discovery without using any student performance data in several ill-structured domains: Rumble Blocks, Chinese Character, and Article Selection. This is especially effective and useful in domains where an accurate human-authored cognitive model is unavailable or authoring a cognitive model is difficult. Secondly, for domains where a cognitive model is available, we show that representations learned through CogRL can be used to get accurate estimates of skill difficulty and learning rate parameters without using any student performance data. These estimates are shown to highly correlate with estimates using student performance data on an Article Selection dataset.

While I did learn a lot of maths while doing my engineering degree, I forgot most of it by the time I wanted to get into Machine Learning. After I graduated I never really had a need for any of the maths. I did a lot of web programming which relied on logic and I can honestly say that with each system with the word'Management' in the title I lost a third of my math knowledge! I've programmed extensions for Learning Management Systems, Content Management Systems and Customer Relationship Management Systems -- I'll leave you to figure out how much math apptitude I had after working with these systems. At the moment I've got good data science skills and can use a variety of ML and DL algorithms.

Dana Bowman, 56, expresses gratitude for fresh produce at least 10 times in the hour and a half we're having coffee on a frigid spring day in Lindsay, Ontario. Over the many years she scraped by on government disability payments, she tended to stick to frozen vegetables. She'd also save by visiting a food bank or buying marked-down items near or past their sell-by date. But since December, Bowman has felt secure enough to buy fresh fruit and vegetables. She's freer, she says, to "do what nanas do" for her grandchildren, like having all four of them over for turkey on Easter.

Starting this fall, some students at Northeastern University in Boston will be given the option of getting an Echo Dot smart speaker linked to their university accounts. They'll be able to ask Amazon's Alexa what time their classes are, how much money's left on their food card and even how much they owe the bursar's office. The program gives students instant access to information they would have to call or go online for, as well as taking pressure off the school's offices. It also makes Amazon's digital assistant a go-to source for a generation who will inhabit a world in which talking to computers is commonplace and who will soon have paychecks to spend. At the same time, it raises questions about security and privacy for young adults living in close quarters, often on their own for the first time.

As AI continues to advance, what future jobs are just over the horizon? Despite fears of automation taking away jobs, the need for skilled humans to operate, utilise and advance technologies will remain unequivocally necessary. While there are plenty of people who fear robots taking over their jobs, there are also many important positives to automation. This starts with having robots in the workplace to treat like robots, and revaluing human employees as actual humans with a need for purpose and work-life balance. Automation, and augmented and virtual reality (AR/VR) all lead to the idea that human workers will be valued for their uniquely human skills, such as creativity and innovative thinking.

In Ann Arbor, Michigan, last week, 125 mostly white, mostly male, business-card-bearing attendees crowded into a brightly lit ballroom to consider "mobility." That's the buzzword for a hazy vision of how tech in all forms--including smartphones, credit cards, and autonomous vehicles-- will combine with the remains of traditional public transit to get urbanites where they need to go. There was a fizz in the air at the Meeting of the Minds session, advertised as a summit to prepare cities for the "autonomous revolution." In the US, most automotive research happens within an hour of that ballroom, and attendees knew that development of "level 4" autonomous vehicles--designed to operate in limited locations, but without a human driver intervening--is accelerating. Susan Crawford (@scrawford) is an Ideas contributor for WIRED, a professor at Harvard Law School, and the author of Captive Audience: The Telecom Industry and Monopoly Power in the New Gilded Age.