We present a straightforward non-iterative method for shallowing of deep Convolutional Neural Network (CNN) by combination of several layers of CNNs with Advanced Supervised Principal Component Analysis (ASPCA) of their outputs. We tested this new method on a practically important case of'friend-or-foe' face recognition. This is the backyard dog problem: the dog should (i) distinguish the members of the family from possible strangers and (ii) identify the members of the family. Our experiments revealed that the method is capable of drastically reducing the depth of deep learning CNNs, albeit at the cost of mild performance deterioration. 1. Introduction IT giants have produced many software "semiproducts" for image recognition. This new opportunity gave rise to many works in face recognition. These works and popular critics of their results prove that the performance of these systems are problem-depending and the devil is in the detail of testing and validation: the systems, which are almost perfect for one problem can be useless for another one. In this paper we focus on a problem which, on the one hand, appears to be a close relative of the face recognition applications and yet, on the other hand, is somewhat more relaxed.

Network structure optimization is a fundamental task in complex network analysis. However, almost all the research on Bayesian optimization is aimed at optimizing the objective functions with vectorial inputs. In this work, we first present a flexible framework, denoted graph Bayesian optimization, to handle arbitrary graphs in the Bayesian optimization community. By combining the proposed framework with graph kernels, it can take full advantage of implicit graph structural features to supplement explicit features guessed according to the experience, such as tags of nodes and any attributes of graphs. The proposed framework can identify which features are more important during the optimization process. We apply the framework to solve four problems including two evaluations and two applications to demonstrate its efficacy and potential applications.

The Multimedia Neural Cognitive Computing (MNCC) model was designed based on the nervous mechanism and cognitive architecture. Furthermore, the semantic-oriented hierarchical Cross-media Neural Cognitive Computing (CNCC) framework was proposed based on MNCC, and formal description and analysis for CNCC was given. It would effectively improve the performance of semantic processing for multimedia information, and has far-reaching significance for exploration and realization brain-inspired computing. Keywords Deep learning·cognitive computing·brain-inspired computing·cross-media neural cognitive computing·multimedia neural cognitive computing 1 Introduction The brain-inspired computing (BIC) is the integration of neural cognitive science and information technology. It would realize state-of-the-art computing system which has advanced in energy consumption, computing ability and efficiency.

In 2017, these 15 self-driving car start-ups have cumulatively raised funding of more than $3 billion till date and the funding raised in 2017 was 100% more than 2016. The total funding of all self-driving scar tart-ups from both private and corporate investors has gone past $5.5 billion as of March 2018.The digitization is also fuelling the prospects of autonomous on demand ride hailing taxis. It is worth mentioning that only few of the start-ups are generating revenue and some are yet to showcase a clear path to scale up and ultimately become profitable. US is considered to be leading the way in terms of legislation for driverless vehicles. States in America including Nevada, Florida, California and Michigan have already passed laws concerning driverless cars.

Online dating apps aren't sitting idly by as Facebook makes its big leap into helping users find love on its platform. At its annual F8 developers conference on Tuesday, CEO Mark Zuckerberg revealed Facebook's plans to add a dating feature to the site. Now, popular dating apps including Tinder, Bumble and Hinge are firing back at the social media giant. Facebook CEO Mark Zuckerberg (pictured) introduced the firm's new dating feature at the firm's annual F8 developers conference in San Jose, California on Tuesday'Come on in, the water's warm. Their product would be great for US/Russia relationships'.

Ford could soon make the futuristic Batmobile a reality. The carmaker recently filed a patent for a concept vehicle that has a detachable motorcycle stored inside it. Batman fans will be psyched to know that the car is similar to the Tumbler driven around by the superhero in'The Dark Knight' trilogy, which can deploy a Batpod motorcycle at a moment's notice. Ford recently filed a patent for a concept vehicle that has a detachable motorcycle stored inside it. The patent describes a'multimodal passenger transportation apparatus,' with the main vehicle appearing to be a Ford Focus hatchback.

ExtraHop machine learning technology is now available globally in both its performance management and Reveal(x) solutions. When paired with the ExtraHop analytics-first workflow, IT teams are instantly alerted to security and performance issues that they can then easily and rapidly investigate from high-level performance metrics to individual transactions to packets – all in a matter of clicks. The broad availability of machine learning within the ExtraHop performance and security products delivers enhanced value to customers by surfacing the most immediate security threats and performance impacts. ExtraHop's advanced machine learning is a highly secure and scalable technology that uses strong detectors based on dimensionality reduction and outlier detection to identify anomalies in real time. The technology builds self-adapting models for every device, network and application, and then proactively detects and surfaces potential issues in the environment directly into the ExtraHop UI.

Do you ever find it easier to describe your preferred haircut using a picture rather than words? That's because your brain is hardwired to excel at visual selection. Neurons devoted to visual processing take up 30% of the cortex, versus 10% combined for touch and hearing. Visual search joins the left side of the brain with the right, allowing for more accurate searches and more creative results. Brands that grab the attention of users at this stage have a better chance of nudging would-be consumers towards purchase.

Ever been delayed on a flight because of straggling fellow passengers? That might be an annoyance of the past at Singapore's Changi airport, which is testing facial recognition systems that could, in future, help locate lost travellers or those spending a little too much time in the duty-free shops. Changi Airport, ranked the world's best for six years straight in a survey by air travel consultancy Skytrax, is looking at how it can use the latest technologies to solve many problems - from cutting taxiing times on the runway to quicker predictions of flight arrivals. It could be used to find late passengers at Singapore's Changi Airport It comes as the island state embarks on a'smart nation' initiative to utilise technology to improve lives, create economic opportunity and build community ties. However the proposed use of cameras mounted on lampposts that are linked to facial recognition software has raised privacy concerns.

Artificial intelligence and machine learning are increasingly automating human tasks, but it's not all bad news for us. In fact, among the first areas where this advanced technology is already adding value is AI marketing and conversational commerce. Intelligent systems are helping businesses more effectively connect customers in important touch points ranging from marketing and sales to customer support. The irony of it is that AI can help make business personal again via "extreme personalization", where systems and their human counterparts, learn to deliver meaningful value to individual customers at scale through each engagement over time. But to do so requires more than intelligent technology, it requires a change in executive mindsets to shift the course of AI from scaling one-to-many customer experiences to that of customer-centered extreme personalization.