"Here" is a funny word. So says Scott Hassan, the media-shy Silicon Valley billionaire who once set out to build the first fully autonomous humanoid robot, and ended up hawking what looks like a flatscreen atop two long legs with wheels. If Hassan has his way, this seemingly simple device--called a Beam--will close the loop between cyberspace and meat-space. The Beam was designed as a video conferencing tool, allowing instant, face-to-face communication--kind of like FaceTime or Skype, except you can drive the screen on legs around the room remotely, with a keyboard. It could one day become much more.

The backpropagation algorithm for calculating gradients has been widely used in computation of weights for deep neural networks (DNNs). This method requires derivatives of objective functions and has some difficulties finding appropriate parameters such as learning rate. In this paper, we propose a novel approach for computing weight matrices of fully-connected DNNs by using two types of semi-nonnegative matrix factorizations (semi-NMFs). In this method, optimization processes are performed by calculating weight matrices alternately, and backpropagation (BP) is not used. We also present a method to calculate stacked autoencoder using a NMF. The output results of the autoencoder are used as pre-training data for DNNs. The experimental results show that our method using three types of NMFs attains similar error rates to the conventional DNNs with BP.

Conversations about basic income, a government-funded salary given to every citizen, used to take place in the dingy offices of extremist left-wing politicians, or in the campus dorm rooms of idealistic students determined to fix the problems of the previous generation. The conversation was about social responsibility. It wasn't an economic case, it was a moral one. A sharp uptake in technology designed to automate jobs and replace human workers is bringing new voices to this old debate. Today's society could be disastrously affected by artificial intelligence and growing automation, and scientists and technologists are looking for ways to stop that damage before it happens.

There are a lot of major problems today with tangible, real-world consequences. A short list might include terrorism, U.S.-Russian relations, climate change and biodiversity loss, income inequality, health care, childhood poverty, and the homegrown threat of authoritarian populism, most notably associated with the presumptive nominee for the Republican Party, Donald Trump. Yet if you've been paying attention to the news for the past several years, you've almost certainly seen articles from a wide range of news outlets about the looming danger of artificial general intelligence, or "AGI." For example, Stephen Hawking has repeatedly expressed that "the development of full artificial intelligence could spell the end of the human race," and Elon Musk -- of Tesla and SpaceX fame -- has described the creation of superintelligence as "summoning the demon." Furthermore, the Oxford philosopher and director of the Future of Humanity Institute, Nick Bostrom, published a New York Times best-selling book in 2014 called Superintelligence, in which he suggests that the "default outcome" of building a superintelligent machine will be "doom." Should we really be worried about a takeover by killer computers hell-bent on the total destruction of Homo sapiens?

The pathway in the brain responsible for vision operates in several layers, each of which is thought to extract progressively more information from an image, such as movement, shape, color, and so on. Each layer consists of huge numbers of neurons connected into a vast network. Deep convolution neural networks have a similar structure. They too are made up of layers, and each of these is a network of circuits designed to mimic the behavior of neurons, hence the term neural network. Through much trial and error, computer scientists have found that these layers perform best when each extracts progressively more information about an image. And when they look at the behavior of layers individually, they find remarkable similarities to the function of specific layers in the brain.

The weekend is upon us once again (woohoo!), and after another week packed with updates, announcements, rumors and insights, its arrival is certainly welcome. And as ever, 7 Days is here too, to walk you through what's been happening and bring you up to speed. We begin our odyssey this week in the UK, where the European Commission has blocked Three's proposed 10.25 billion acquisition of O2 UK, which would have created the country's largest mobile network operator. The EC determined that the deal would harm innovation, limit competition, and increase prices across the market. Cloud CRM platform Salesforce suffered an outage earlier this week affecting a relatively small proportion of its customers.

This origami robot is designed to be swallowed in an ice capsule (left) before unfolding inside the stomach. Scientists have created a tiny, ingestible robot made from pig intestine that might be able to retrieve swallowed objects without surgery. The robot is built from a magnet attached to folds of dried meat typically used in sausage casings. It needs no tether or power source, instead being guided by magnets outside the body. Though swallowed in an ice capsule, the robot unfolds inside the stomach as its casing melts.

As you read this, some 100 billion neurons are transmitting information through electrical and chemical signals via synapses in your brain. Given the central role these cells play in neurological functioning, it's perhaps not surprising they typically hog the limelight -- after all, the signals they transmit lie at the heart of human behavior, from the simplest of movements to the most complex of thoughts. It's worth noting, however, that complimentary cells called astrocytes actually outnumber neurons in the brain. Unfortunately, these star-shaped cells have largely been ignored in neurological research because they don't fire electrical impulses in the same way that neurons do. Yukiko Goda of the Riken Brain Science Institute in Saitama is helping to correct this. Goda, who was born in Osaka, went to high school in Toronto after her father was transferred to Canada for work.

Today our focus has been on KPIs, ERP, WMS, TMS, YMS, EDI, The Cloud, S and OP, 3 D Printing, IoT, IoE, Drones: Same Hour/Day/Time Delivery to Customers, Cyber Security, Theft, Government Regulations, E-Commerce, Omni-Channel, Modeling/Simulation, Risk Management, Tracking, Traceability, Re-shoring, Robotics, et al, but…what about Artificial Intelligence or AI technologies? AI is a controversy of deep, lasting dimensions. Will machines learn to think like humans…and then outthink us? The application of AI technologies has created the ability to understand, store and use product information in an entirely new way. AI technologies allow you to understand the underlying grammatical structure, the product DNA, used to build each product, together with all natural variations that potentially exist.

When the tech giant Microsoft unveiled its AI or artificial intelligence-powered bot on Twitter for a playful chat with the people in March, little did the tech giant realise that the twitterati would begin slamming the innocent bot with racist and offensive comments. Launched as an experiment in "conversational understanding" and to engage people through "casual and playful conversation", Tay was soon taken off Twitter by Microsoft engineers. This was a soft experiment. But what if you can interact with a "chatbot" and send the AI-powered machine your financial requirements like you would text to your banker or chartered accountant in the near future? Facebook wants this to happen and at its F8 global developer conference in April, the social networking giant unveiled AI bots right into its popular messaging app Messenger -- to allow 900 million monthly active users on Messenger to interact with businesses and get updates from them.