Posted in Elon Musk, robotics/AI
When you use smartphone AI apps like Siri, you’re dependent on the cloud for a lot of the processing — limited by your connection speed. But what if your smartphone could do more of the processing directly on your device — allowing for smarter, faster apps?
MIT scientists have taken a step in that direction with a new way to enable artificial-intelligence systems called convolutional neural networks (CNNs) to run locally on mobile devices. (CNN’s are used in areas such as autonomous driving, speech recognition, computer vision, and automatic translation.) Neural networks take up a lot of memory and consume a lot of power, so they usually run on servers in the cloud, which receive data from desktop or mobile devices and then send back their analyses.
Forget swiping a credit card or badge to buy food at work. One Wisconsin-based tech firm is offering to install rice-size microchips in its employees’ hands.
Three Square Market will be the fir st firm in the U.S. to use the device, which was approved by the FDA in 2004, CEO Todd Westby told CNBC on Monday.
“We think it’s the right thing to do for advancing innovation just like the driverless car basically did in recent months,” he said in an interview with “Closing Bell.”
The unique swimming strategies of natural microorganisms have inspired recent development of magnetic micro/nanorobots powered by artificial helical or flexible flagella. However, as artificial nanoswimmers with unique geometries are being developed, it is critical to explore new potential modes for kinetic optimization. For example, the freestyle stroke is the most efficient of the competitive swimming strokes for humans. Here we report a new type of magnetic nanorobot, a symmetric multilinked two-arm nanoswimmer, capable of efficient “freestyle” swimming at low Reynolds numbers. Excellent agreement between the experimental observations and theoretical predictions indicates that the powerful “freestyle” propulsion of the two-arm nanorobot is attributed to synchronized oscillatory deformations of the nanorobot under the combined action of magnetic field and viscous forces. It is demonstrated for the first time that the nonplanar propulsion gait due to the cooperative “freestyle” stroke of the two magnetic arms can be powered by a plane oscillatory magnetic field. These two-arm nanorobots are capable of a powerful propulsion up to 12 body lengths per second, along with on-demand speed regulation and remote navigation. Furthermore, the nonplanar propulsion gait powered by the consecutive swinging of the achiral magnetic arms is more efficient than that of common chiral nanohelical swimmers. This new swimming mechanism and its attractive performance opens new possibilities in designing remotely actuated nanorobots for biomedical operation at the nanoscale.
Each bot is 5 micrometres long and has three main parts, connected together like sausage links by two silver hinges. Its gold body is flanked by two magnetic arms made of nickel, and applying a magnetic field to the tiny robot makes the arms move.
The next generation bloodstream will be made from biodegradable materials before they can be used in the bloodstream. Less complicated areas in the human body like the urinary tract or the eyeballs should see clinical trials begin within the next five to 10 years. Injecting a single swimmer into an eyeball, where it could deliver medication directly to the retina and then be removed, would be much less complicated than letting a swarm of them swim throughout the entire circulatory system.
Is this the beginning of Skynet? A robot is being hired to help out shoppers at a St. Louis grocery store chain.
Tally will be working the aisles at Schnucks grocery stores looking for items that are out of stock and checking on prices, the St. Louis Post-Dispatch reported.
Google’s DeepMind has revealed a radical new research project designed to give AI’s an imagination.
The breakthrough means that systems will be able to think about their actions, and undertake ‘deliberate reasoning.’
The radical system uses an internal ‘imagination encoder’ that helps the AI decide what are and what aren’t useful predictions about its environment.
By Marc Pollefeys, Director of Science, HoloLens
It is not an exaggeration to say that deep learning has taken the world of computer vision, and many other recognition tasks, by storm. Many of the most difficult recognition problems have seen gains over the past few years that are astonishing.
Although we have seen large improvements in the accuracy of recognition as a result of Deep Neural Networks (DNNs), deep learning approaches have two well-known challenges: they require large amounts of labelled data for training, and they require a type of compute that is not amenable to current general purpose processor/memory architectures. Some companies have responded with architectures designed to address the particular type of massively parallel compute required for DNNs, including our own use of FPGAs, for example, but to date these approaches have primarily enhanced existing cloud computing fabrics.
Hot on the heels of last month’s nuclear fusion breakthrough comes the first results from a multi-year partnership between Google and Tri Alpha Energy, the world’s largest private fusion company. The two organizations joined forces in 2014 in the hopes that Google’s machine learning algorithms could advance plasma research and bring us closer to the dream of fusion power.
