The heavy-tailed distributions of corrupted outliers and singular values of all channels in low-level vision have proven effective priors for many applications such as background modeling, photometric stereo and image alignment. And they can be well modeled by a hyper-Laplacian. However, the use of such distributions generally leads to challenging non-convex, non-smooth and non-Lipschitz problems, and makes existing algorithms very slow for large-scale applications. Together with the analytic solutions to lp-norm minimization with two specific values of p, i.e., p=1/2 and p=2/3, we propose two novel bilinear factor matrix norm minimization models for robust principal component analysis. We first define the double nuclear norm and Frobenius/nuclear hybrid norm penalties, and then prove that they are in essence the Schatten-1/2 and 2/3 quasi-norms, respectively, which lead to much more tractable and scalable Lipschitz optimization problems. Our experimental analysis shows that both our methods yield more accurate solutions than original Schatten quasi-norm minimization, even when the number of observations is very limited. Finally, we apply our penalties to various low-level vision problems, e.g., text removal, moving object detection, image alignment and inpainting, and show that our methods usually outperform the state-of-the-art methods.

In this work, we study a new approach to optimizing the margin distribution realized by binary classifiers. The classical approach to this problem is simply maximization of the expected margin, while more recent proposals consider simultaneous variance control and proxy objectives based on robust location estimates, in the vein of keeping the margin distribution sharply concentrated in a desirable region. While conceptually appealing, these new approaches are often computationally unwieldy, and theoretical guarantees are limited. Given this context, we propose an algorithm which searches the hypothesis space in such a way that a pre-set "margin level" ends up being a distribution-robust estimator of the margin location. This procedure is easily implemented using gradient descent, and admits finite-sample bounds on the excess risk under unbounded inputs. Empirical tests on real-world benchmark data reinforce the basic principles highlighted by the theory, and are suggestive of a promising new technique for classification.

Low-rank tensor decomposition is a promising approach for analysis and understanding of real-world data. Many such analyses require correct recovery of the true latent factors, but the conditions of exact recovery are not known for many existing tensor decomposition methods. In this paper, we derive such conditions for a general class of tensor decomposition methods where each latent tensor component can be reshuffled into a low-rank matrix of arbitrary shape. The reshuffling operation generalizes the traditional unfolding operation, and provides flexibility to recover true latent factors of complex data-structures. We prove that exact recovery can be guaranteed by using a convex program when a type of incoherence measure is upper bounded. The results on image steganography show that our method obtains the state-of-the-art performance. The theoretical analysis in this paper is expected to be useful to derive similar results for other types of tensor-decomposition methods.

If you think regular snakes are scary, try a robotic serpent that can scale ladders. Japanese researchers at Kyoto University have unveiled a prototype'robot snake' that can slither and curl around surfaces and climb just like the real thing. Rather than just creating it to trigger nightmares, they believe it could one day be used to save lives. Kyoto University researchers say the robot snake could be used to enter dangerous situations that are unsafe for humans. It may also be able to rescue humans that are stuck in hard to reach places.

Microsoft and on-demand transportation and mobile payments company Grab have formed a strategic partnership to transform the delivery of digital services and mobility in Southeast Asia. The venture plans to leverage Microsoft's machine learning and artificial intelligence capabilities and Grab will adopt Microsoft Azure as its preferred cloud platform.

AI and related technologies are most effective when used as a way to unleash creativity and increase autonomy in workers. The study was done by researchers at Goldsmiths, University of London, and commissioned by robotic process automation vendor Automation Anywhere. Released in September, the study looked at the role and impact of automation in the workplace, focusing on two "augmentation" technologies: Among the findings: Not only do enterprises that invest in augmented human intelligence promote a "more human workplace," but workplaces with cultures that foster learning also do technology augmentation more successfully. "The key [to success] was to invest in technology and people," said Mihir Shukla, CEO of Automation Anywhere. "Technology investment is straightforward," he added.

Google is secretly planning to launch a censored version of its search engine in China within the next year, a leaked transcript seems to reveal. According to The Intercept, Google's search engine chief Ben Gomes held a meeting in July to discuss the progress of a new search engine, dubbed Project Dragonfly. The platform would blacklist words and phrases like "human rights," "Nobel Prize," and "student protest," in order to conform with China's strict censorship laws. "You have taken on something extremely important to the company," Mr Gomes told the Google employees, according to the transcript obtained by the publication. "I have to admit it has been a difficult journey. But I do think a very important and worthwhile one. And I wish ourselves the best of luck in actually reaching our destination as soon as possible."

Autonomous cars are being programmed to interpret road traffic and pedestrians in order to drive more like humans. A key issue with driverless cars has been their ability to interpret traffic and other upcoming object such as pedestrians, where they become overly cautious. For example, if a person appears as if they will cross the street but then changes their mind, a driverless vehicle may stop and wait. Engineers at Perceptive Automata - based near Boston - has teamed up with Hyundai Cradle, the car firm's technology investment arm, to create software that anticipates what pedestrians, cyclists and other motorists might do. The newly-developed artificial software will then help the driverless vehicles predicate what is coming up more like a human.

Artificial Intelligence and Machine Learning have barged into our lives without our knowledge and have now taken a permanent spot in there. Despite the scare of machines taking over human jobs, the technology has worked in human kind's favour majorly. The innovation of AI and ML has indeed made our lives earlier and convenient. While the industry is at a nascent stage in India, people are aware of their presence. Every day these innovations are contributing to improving people's way of living along with increasing their level of productivity.

Softbank Robotics today announced that its robot Pepper will now use emotion recognition AI from Affectiva to interpret and respond to human activity. Pepper is about four feet tall, gets around on wheels, and has a tablet in the center of its chest. The humanoid robot made its debut in 2015 and was designed to interact with people. Cameras and microphones are used to help Pepper recognize human emotions, like hostility or joy, and respond appropriately with a smile or indications of sadness. This type of intelligence likely comes in handy for the environments where Pepper operates, like banks, hotels, and Pizza Huts in some parts of Asia.