I've always had a lot of respect for veterinarians, because they are masters at solving problems based purely on fuzzy symptoms that their patients cannot explain: where it hurts, how long it's been hurting, and what events led up to the problem. Many times the patients don't even know they are sick. Yet a vet is able to make educated guesses with the data they do have, which often results in successful diagnoses and treatments. Wireless local area networks (WLANs) cannot talk, either, which often forces IT administrators to operate like doctors, listening to wireless users describe symptoms in vague terms: "I can't connect." "Sometimes it works, sometimes it doesn't."

We study the statistical limits of both detecting and estimating a rank-one deformation of a symmetric random Gaussian tensor. We establish upper and lower bounds on the critical signal-to-noise ratio, under a variety of priors for the planted vector: (i) a uniformly sampled unit vector, (ii) i.i.d. $\pm 1$ entries, and (iii) a sparse vector where a constant fraction $\rho$ of entries are i.i.d. $\pm 1$ and the rest are zero. For each of these cases, our upper and lower bounds match up to a $1+o(1)$ factor as the order $d$ of the tensor becomes large. For sparse signals (iii), our bounds are also asymptotically tight in the sparse limit $\rho \to 0$ for any fixed $d$ (including the $d=2$ case of sparse PCA). Our upper bounds for (i) demonstrate a phenomenon reminiscent of the work of Baik, Ben Arous and P\'ech\'e: an `eigenvalue' of a perturbed tensor emerges from the bulk at a strictly lower signal-to-noise ratio than when the perturbation itself exceeds the bulk; we quantify the size of this effect. We also provide some general results for larger classes of priors. In particular, the large $d$ asymptotics of the threshold location differs between problems with discrete priors versus continuous priors. Finally, for priors (i) and (ii) we carry out the replica prediction from statistical physics, which is conjectured to give the exact information-theoretic threshold for any fixed $d$. Of independent interest, we introduce a new improvement to the second moment method for contiguity, on which our lower bounds are based. Our technique conditions away from rare `bad' events that depend on interactions between the signal and noise. This enables us to close $\sqrt{2}$-factor gaps present in several previous works.

Their work is supported in part by the National Science Foundation under Grant No. 1254204 and the Office of Naval Research Grant No. N00014-15-1-2677. S. Rangan (email: srangan@nyu.edu) is with the Department of Electrical and Computer Engineering, New York University, Brooklyn, NY. His work was supported by the National Science Foundation under Grant No. 1116589 and the industrial affiliates of NYU WIRELESS. P. Schniter (email: schniter@ece.osu.edu) is with the Department of Electrical and Computer Engineering, The Ohio State University. His work was supported in part by the National Science Foundation under Grants CCF-1218754 and CCF-1527162.

With 1 to 10 Gigabytes per second bandwidth; 1 millisecond latency and supporting a density of 100 or more devices in any given room size location, it is truly instant, always-on, and able to download the equivalent of a whole movie in a few seconds. This is the new competitive level of performance for the 21st century economy. It enables real-time mission critical rapid response such as connected self-driving cars to enabling the Internet of things sensors and mobile devices that will drive connected buildings, smart cities, savings energy, enabled new skills and services, changing how markets and work gets done.

Techies around the world would have been hard-wired for insights last week as the DLD17 conference in Munich and the World Economic Forum (WEF) in Davos ran concurrently. The former – populated by academics and innovators – was marked by a frisky anticipation over technological possibilities, while the latter – made up of politicians and business leaders – succumbed to fear and uncertainty over global events, even as the sun shone outside. Chinese premier, Xi Jinping, speaking in Davos, said: "Unlike the industrial revolution, our fourth technological revolution is unfolding at an exponential, and not linear, pace". One statistic that came out of the meeting was how Hollywood box office takings have been outpaced by App revenues – even though they didn't exist before 2008. Premier Xi opened his address with a quote from Charles Dickens: "It was the best of times, it was the worst of times", to highlight the many contradictions facing modern society.

A Northwestern University team developed a new computational model that performs at human levels on a standard intelligence test. This work is an important step toward making artificial intelligence systems that see and understand the world as humans do. "The model performs in the 75th percentile for American adults, making it better than average," said Northwestern Engineering's Ken Forbus. "The problems that are hard for people are also hard for the model, providing additional evidence that its operation is capturing some important properties of human cognition." The new computational model is built on CogSketch, an artificial intelligence platform previously developed in Forbus' laboratory.

Theresa May is due to announce details of the "Modern Industrial Strategy" at a regional meeting of the cabinet in the North West. The new plan will center on ten key strategic pillars, according to the release unveiled by the Prime Minister on Sunday. "We must become a more innovative economy and do more to commercialize our world-leading science base to drive growth across the UK," reads the first point, called "Investing in science, research, and innovation." The government is going to spend £4.7 billion ($5.85 billion) on the new strategy with the money steered to such areas as AI, "smart" energy technology, robotics, and 5G wireless. The project will improve living standards and drive economic growth across the whole country, according to Business and Energy Secretary Greg Clark.

LONDON -- The government is putting cutting edge technologies such as artificial intelligence (AI), 5G wireless internet, "smart" energy technology, and robotics at the heart of its new post-Brexit industrial strategy. Theresa May is set to launch the government's "Modern Industrial Strategy" on Monday at a regional meeting of the cabinet in the North West. The Prime Minister announced in a release on Sunday evening that the strategy would be focused around ten key strategic pillars, the first of which is: "Investing in science, research, and innovation." The government says that a £4.7 billion increase in research & development (R&D) funding announced in last year's Autumn Statement is central to the new industrial strategy. This investment will go to areas such as AI, "smart" energy technology, robotics, and 5G wireless, the only specific industries name-checked in the release.

Australia has started implementing biometric facial, iris and fingerprint recognition in airports, allowing passengers to go through without showing a passport or even talking to anyone. The "Seamless Traveler" project is aimed at creating a "fast, seamless self-processing experience for up to 90 percent of travelers," so that border control can focus on high-risk passengers. The handy, but invasive-sounding plan would allow international travelers to "literally just walk out like at a domestic airport," security analyst John Coyne told Australia's Sidney Morning Herald. The system would replace passport-scanning SmartGates, which were implemented in the nation just ten years ago. The government's plan to implement biometrics might be a touch ambitious, however.

Tiny drones could scout high-rise buildings and underground tunnels for possible threats to US troops in cities of the future. But instead of spending years cooking up the necessary drone technologies in military research labs, the Pentagon might be better off shopping for the latest civilian drones coming soon to stores. US military leaders have discussed the need for a new generation of scout drones for some time. After all, kicking down doors is a dirty and dangerous business for US troops trying to clear enemy-held buildings. It would be far safer to deploy diminutive drone buddies to provide an initial peek inside, and identify any potential threats.