Goto

Collaborating Authors

 Asia


Learning Deep Features via Congenerous Cosine Loss for Person Recognition

arXiv.org Machine Learning

Person recognition aims at recognizing the same identity across time and space with complicated scenes and similar appearance. In this paper, we propose a novel method to address this task by training a network to obtain robust and representative features. The intuition is that we directly compare and optimize the cosine distance between two features - enlarging inter-class distinction as well as alleviating inner-class variance. We propose a congenerous cosine loss by minimizing the cosine distance between samples and their cluster centroid in a cooperative way. Such a design reduces the complexity and could be implemented via softmax with normalized inputs. Our method also differs from previous work in person recognition that we do not conduct a second training on the test subset. The identity of a person is determined by measuring the similarity from several body regions in the reference set. Experimental results show that the proposed approach achieves better classification accuracy against previous state-of-the-arts.


Hankel Matrix Nuclear Norm Regularized Tensor Completion for $N$-dimensional Exponential Signals

arXiv.org Machine Learning

Signals are generally modeled as a superposition of exponential functions in spectroscopy of chemistry, biology and medical imaging. For fast data acquisition or other inevitable reasons, however, only a small amount of samples may be acquired and thus how to recover the full signal becomes an active research topic. But existing approaches can not efficiently recover $N$-dimensional exponential signals with $N\geq 3$. In this paper, we study the problem of recovering N-dimensional (particularly $N\geq 3$) exponential signals from partial observations, and formulate this problem as a low-rank tensor completion problem with exponential factor vectors. The full signal is reconstructed by simultaneously exploiting the CANDECOMP/PARAFAC structure and the exponential structure of the associated factor vectors. The latter is promoted by minimizing an objective function involving the nuclear norm of Hankel matrices. Experimental results on simulated and real magnetic resonance spectroscopy data show that the proposed approach can successfully recover full signals from very limited samples and is robust to the estimated tensor rank.


What If We Had Perfect Robot Referees?

The New Yorker

Last month, Mark Clattenburg, who is generally regarded as one of the finest referees in professional soccer, left England's Premier League for a better-paying position in Saudi Arabia. Plenty of famous players have chosen riches over prestige and joined less established leagues in Asia and the Middle East. But this was one of the first times that a referee of Clattenburg's stature and prominence had left in his prime. Just last year, he was selected to referee two of the most important matches on Earth: the finals of the Champions League, in May, and then the European Championship, two months later--plum gigs that testified to his skill. To commemorate the occasion, he had the feat tattooed on his arm--a testament to his knack for preening.


Toyota to use artificial intelligence in search for new batteries

#artificialintelligence

Toyota Motor Corp will use artificial intelligence in new research to speed up the discovery of advanced battery materials and fuel-cell catalysts to power electric and other emission-free vehicles, the company said on Thursday. Toyota will invest around $35 million in its North American research arm, the Toyota Research Institute (TRI) and will involve collaboration with a number of U.S. academic institutions including the University of Michigan and the Massachusetts Institute of Technology, and British material sciences company Ilika. Artificial intelligence will be used to identify new possible materials for batteries and fuel and run computer tests to narrow down the field for simulation tests by researchers. Automakers have been investing heavily in developing new batteries and fuel cells to increase the range of electric vehicles. China in particular is banking on the heavy use of electric vehicles in the future.


For Google, the AI Talent Race Leads Straight to Canada

WIRED

America's biggest tech companies are remaking the internet through artificial intelligence. And more than ever, these companies are looking north to Canada for the ideas that will advance AI itself. This morning, Google announced it's starting an AI lab in Toronto. At the same time it's helping to fund a public-private partnership with the University of Toronto to develop and commercialize AI talent and ideas. In November, the company made a similar move in Montreal--a city that has also attracted Microsoft's attention.


FOX NEWS HALFTIME REPORT: Trump's re-election bid collides with policy problems

FOX News

On the roster: Trump's re-election bid collides with policy problems - Witness: Russians targeted Rubio - Report: Trump aides told Nunes what Nunes told Trump - I'll Tell You What: Kidding, not kidding - Paging Tara Reid TRUMP'S RE-ELECTION BID COLLIDES WITH POLICY PROBLEMS It's too soon to say how President Trump's agenda will fair, but we do know his re-election bid is in trouble. That may sound preposterous to say in the 10th week of an administration, but here we are. Trump, who filed for re-election before he took office, is getting a boost from his most important donors, hedge-fund tycoon Robert Mercer and his family. It comes in the form of a $1.3 million ad blitz targeted at swing states as well as states represented by vulnerable Democratic senators. Trump is doing his part by renewing his war with his fellow Republicans, blasting House conservatives for defeating his health-insurance overhaul last week.


Fukushima News: Evacuation Zone Around Nuclear Power Plant Reduced

International Business Times

The Japanese government Friday eased evacuation orders for towns not seriously contaminated by the 2011 Fukushima Daiichi nuclear power plant disaster. The government lifted evacuation orders for parts of Kawamata, Namie and Iitate, Kyodo News reported. The order also frees a large part of Tomioka on Saturday. The action reduces the evacuation zones by two-thirds but it was unclear whether residents actually would return to their homes because of radiation fears and a lack of amenities like schools. The most seriously contaminated areas remain off-limits.


What is AI? Even Elon Musk Can't Explain

#artificialintelligence

Word leaked Monday via The Wall Street Journal that Tesla / SpaceX industrialist Elon Musk has been funding a company called Neuralink-- allegedly with some of his own money -- attempting to connect computers directly into human brains. This is the same Musk profiled in this month's Vanity Fair, where he tells journalist Maureen Dowd in all seriousness that humanity needs a Mars colony to which we can escape'if AI goes rogue and turns on humanity." In short, Musk is one of many big thinkers who believe a human-computer hybrid is essential to allowing humans to keep their own machines from marginalizing them. Neuralink's technology is said to be a neural lace, which Musk has spoken about for over a year. But for most people, the first question isn't whether artificial intelligence will usurp our planet.


Robots, drones help Japan's builders out of labour crunch Today

Robohub

Construction sites in Japan are enjoying a wave of automation amid an increasing shortage of labourers, with the introduction of robots to do the heavy lifting and drones that fly from above and instantly collect data.


LastPass password manager suffers 'major' security problem

The Independent - Tech

LastPass users are being advised to avoid the password manager while it addresses a "unique and highly sophisticated" security issue. The popular service designed to help internet users protect their online accounts and, as such, is an obvious target for cybercriminals. LastPass hasn't revealed any further details about the problem, but Google's Project Zero security researcher Tavis Ormandy, who discovered it, says it's a serious one. The I.F.O. is fuelled by eight electric engines, which is able to push the flying object to an estimated top speed of about 120mph. The giant human-like robot bears a striking resemblance to the military robots starring in the movie'Avatar' and is claimed as a world first by its creators from a South Korean robotic company Waseda University's saxophonist robot WAS-5, developed by professor Atsuo Takanishi and Kaptain Rock playing one string light saber guitar perform jam session A man looks at an exhibit entitled'Mimus' a giant industrial robot which has been reprogrammed to interact with humans during a photocall at the new Design Museum in South Kensington, London Electrification Guru Dr. Wolfgang Ziebart talks about the electric Jaguar I-PACE concept SUV before it was unveiled before the Los Angeles Auto Show in Los Angeles, California, U.S The Jaguar I-PACE Concept car is the start of a new era for Jaguar.