With the recent renaissance of deep convolution neural networks, encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient training data and fully annotated training data. However, to scale the recognition to a large number of classes with few or now training samples for each class remains an unsolved problem. One approach to scaling up the recognition is to develop models capable of recognizing unseen categories without any training instances, or zero-shot recognition/ learning. This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, and from datasets and evaluation settings. We also overview related recognition tasks including one-shot and open set recognition which can be used as natural extensions of zero-shot recognition when limited number of class samples become available or when zero-shot recognition is implemented in a real-world setting. Importantly, we highlight the limitations of existing approaches and point out future research directions in this existing new research area.

The resemblance between the methods used in studying quantum-many body physics and in machine learning has drawn considerable attention. In particular, tensor networks (TNs) and deep learning architectures bear striking similarities to the extent that TNs can be used for machine learning. Previous results used one-dimensional TNs in image recognition, showing limited scalability and a high bond dimension. In this work, we train two-dimensional hierarchical TNs to solve image recognition problems, using a training algorithm derived from the multipartite entanglement renormalization ansatz (MERA). This approach overcomes scalability issues and implies novel mathematical connections among quantum many-body physics, quantum information theory, and machine learning. While keeping the TN unitary in the training phase, TN states can be defined, which optimally encodes each class of the images into a quantum many-body state. We study the quantum features of the TN states, including quantum entanglement and fidelity. We suggest these quantities could be novel properties that characterize the image classes, as well as the machine learning tasks. Our work could be further applied to identifying possible quantum properties of certain artificial intelligence methods.

The front-line of the stuggle to take out ISIS' most deadly fighters and recruiters like the infamous'White Widow' Sally Jones isn't waged on the battle fields of Syria or Iraq but from an air-conditioned trailer in the middle of the Nevada desert. Seven thousand miles from the war in Syria, pilots sit in a sun-bleached desert base 45 miles north of the bright lights of Las Vegas guiding the lethal Predator drones that have revolutionized modern combat. From cool, dark booths at Creech Air Force Base in Indian Springs, Nevada, pilots and sensor operators work closely with large teams of intelligence analysts who sift streams of real-time data transmitted by the drones on the other side of the planet. The front-line of the stuggle to take out ISIS' most deadly fighters and recruiters isn't waged on the battle fields of Syria or Iraq but from an air-conditioned trailer (pictured) in the middle of the Nevada desert And it is a pilot at this base who is said to have remotely taken out the British ISIS recruiter Sally Jones. Jones, from Chatham in Kent, fled to Syria with her son in 2013 before becoming a recruiter for the terror group.

The power of Algorithms to calculate, contemplate and anticipate the needs of patients is improving rapidly and still has no sign of slowing down. Everything from patient diagnosis to therapy selection will soon be moving at exponential rates. Does that mean the end of doctors? To better understand technology's ever-growing role in healthcare, we first have to better examine the potential of tools and timelines that we are working with. A recent study done at Beth Israel Deaconess Medical Center (BIDMC) and Harvard Medical School, showed that AI isn't about Humans versus Machines. They trained a Deep Learning Algorithm for identifying Metastatic Breast Cancer, interpreting pathology images.

HANGZHOU, China--(BUSINESS WIRE)--Alibaba Group Holding Ltd. ("Alibaba Group") announced today the launch of an innovative global research program, "Alibaba DAMO Academy ("Academy")," which is designed to increase technological collaboration worldwide, advance the development of cutting-edge technology and strive to make the world more inclusive by narrowing the technology gap. With the setup of the Academy, the company expects to invest more than US$15billion in research and development over the next three years. The Academy, which stands for the "Academy for Discovery, Adventure, Momentum and Outlook," will oversee the opening of research and development labs worldwide and seek to recruit talented scientists and researchers to join the program. Alibaba Group's Chief Technology Officer, Jeff ZHANG will be the head of the Academy. In the beginning, the Academy will focus on the opening of seven research labs in China (Beijing and Hangzhou), the United States (San Mateo and Bellevue), Russia (Moscow), Israel (Tel Aviv) and Singapore.

The largest single-dish radio telescope in the world has already begun to spot remarkable objects in the Milky Way, just a year into its career. China's Five hundred-meter Aperture Spherical radio Telescope (FAST) has detected two rapidly rotating stars, known as pulsars, according to the National Astronomical Observatories of China (NAOC). These objects can act like'cosmic clocks' when they spin at a steady rate, and can even shed light on phenomena such as gravitational waves. China's Five hundred-meter Aperture Spherical radio Telescope (FAST) has detected two rapidly rotating stars, known as pulsars, according to the National Astronomical Observatories of China (NAOC). An artist's impression is pictured Pulsars are rotating, highly magnetised neutron stars.

Samsung's Bixby digital assistant hasn't been received well by Galaxy S8, S8 and Note 8 users. And now Samsung is reportedly replacing the head of the Bixby team and might even launch a new version of the assistant next week. Samsung's Rhee In-jong is being replaced as the head of development of the Bixby digital voice assistant by Chung Eui-seok, the head of service intelligence at the company's South Korean headquarters, according to The Investor. The Chung's service group is in charge of developing Bixby and other services related to artificial intelligence. Chung worked for Ericsson before joining Samsung in 2011 and is credited for being the software specialist behind Samsung's security program Knox and Samsung Pay.

Yet another report, this time from Nikkei, claims Apple is facing production issues with the iPhone X's 3D sensors for the new Face ID feature. This could severely impact the availability of the iPhone X on its release date as pre order for the phone are set to begin. The dot projector, dubbed Romeo, uses a laser to beam 30,000 infrared dots on the user's face to map their unique traits, while the receiving module is called Juliet. Trouble with Romeo modules are reportedly holding back mass production of the iPhone X. The report comes as Apple gears up to start taking pre-orders for the iPhone X this month.

Watch: This may be the first footage of a blue whale "heat run." Blue whales are the largest animals on Earth, but we know surprisingly little about their complex social interactions--and they've rarely been recorded on camera. But new footage filmed off the coast of Sri Lanka by pro whale photographer Patrick Dykstra, in conjunction with blue whale scientist Howard Martenstyn, may be a first. Their video shows what they believe is the first known clip of a blue whale "heat run." Heat runs have been well documented in humpback whales, but no known footage exists of the behavior in blue whales (or at least that Dykstra or National Geographic could find).

Comics are big business in Japan, but here in the West, Japanese and American titles alike tend to get overshadowed by movies, television and video games. In fact, many of those programs might even be adaptations of popular comic titles. For its first big VR project, Square Enix's Advanced Technology Division is putting the spotlight back on manga. Due for release in 2018 on all major VR platforms, Project Hikari aims to capture the look and feel of reading a manga while taking advantage of the immersive nature of VR to let the viewer delve deeper into these worlds. Square Enix is best known for console role-playing games like Final Fantasy and Dragon Quest. But it has also been a manga publisher for several decades now, putting out popular titles like Soul Eater, Black Butler and Fullmetal Alchemist.