Lifeboat Foundation

In the past 10 years, the best-performing artificial-intelligence systems—such as the speech recognizers on smartphones or Google’s latest automatic translator—have resulted from a technique called “deep learning.”

Deep learning is in fact a new name for an approach to artificial intelligence called neural networks, which have been going in and out of fashion for more than 70 years. Neural networks were first proposed in 1944 by Warren McCullough and Walter Pitts, two University of Chicago researchers who moved to MIT in 1952 as founding members of what’s sometimes called the first cognitive science department.

Neural nets were a major area of research in both neuroscience and computer science until 1969, when, according to computer science lore, they were killed off by the MIT mathematicians Marvin Minsky and Seymour Papert, who a year later would become co-directors of the new MIT Artificial Intelligence Laboratory.

Hyper-connectivity has changed the way we communicate, wait, and productively use our time. Even in a world of 5G wireless and “instant” messaging, there are countless moments throughout the day when we’re waiting for messages, texts, and Snapchats to refresh. But our frustrations with waiting a few extra seconds for our emails to push through doesn’t mean we have to simply stand by.

To help us make the most of these “micro-moments,” researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a series of apps called “WaitSuite” that test you on vocabulary words during idle moments, like when you’re waiting for an instant message or for your phone to connect to WiFi.

Continue reading “Learn a language while you wait for WiFi” »

Drawing inspiration from the plant world, researchers have invented a new electrode that could boost our current solar energy storage by an astonishing 3,000 percent.

The technology is flexible and can be attached directly to solar cells — which means we could finally be one step closer to smartphones and laptops that draw their power from the Sun, and never run out.

A major problem with reliably using solar energy as a power source is finding an efficient way to store it for later use without leakage over time.

Continue reading “This New Graphene-Based Electrode Could Boost Solar Storage” »

I’ve been reading about Gcam, the Google X project that was first sparked by the need for a tiny camera to fit inside Google Glass, before evolving to power the world-beating camera of the Google Pixel. Gcam embodies an atypical approach to photography in seeking to find software solutions for what have traditionally been hardware problems. Well, others have tried, but those have always seemed like inchoate gimmicks, so I guess the unprecedented thing about Gcam is that it actually works. But the most exciting thing is what it portends.

I think we’ll one day be able to capture images without any photographic equipment at all.

Continue reading “The endgame for cameras is having no camera at all” »

Microscopically fine conductor paths are required on the surfaces of smartphone touchscreens. At the edges of the appliances, these microscopic circuit paths come together to form larger connective pads. Until now, these different conductive paths had to be manufactured in several steps in time-consuming processes. With photochemical metallization, this is now possible in one single step on flexible substrates. The process has several benefits: It is fast, flexible, variable in size, inexpensive and environmentally friendly. Additional process steps for post-treatment are not necessary.

For the new process, the foils are coated with a photoactive layer of metal oxide nanoparticles. “After that, we apply a colorless, UV-stable silver compound,” Peter William de Oliveira, head of optical materials, explains. By irradiation of this sequence of layers, the silver compound disintegrates on the photoactive layer and the silver ions are reduced to form metallic, electrically conductive silver. In this way, paths of varying sizes down to the smallest size of a thousandth of a millimeter can be achieved.

This basic principle allows conductive paths to be created individually. “There are different possibilities we can use depending on the requirements: Writing conductive paths using UV lasers is particularly suitable for the initial customized prototype manufacture and testing a new design of the conductive path. However, for mass production, this method is too time-consuming,” de Oliveira explains.

Continue reading “Silver Circuits On Foil Allow Curved Touchscreens” »

Four years ago, Google started to see the real potential for deploying neural networks to support a large number of new services. During that time it was also clear that, given the existing hardware, if people did voice searches for three minutes per day or dictated to their phone for short periods, Google would have to double the number of datacenters just to run machine learning models.

The need for a new architectural approach was clear, Google distinguished hardware engineer, Norman Jouppi, tells The Next Platform, but it required some radical thinking. As it turns out, that’s exactly what he is known for. One of the chief architects of the MIPS processor, Jouppi has pioneered new technologies in memory systems and is one of the most recognized names in microprocessor design. When he joined Google over three years ago, there were several options on the table for an inference chip to churn services out from models trained on Google’s CPU and GPU hybrid machines for deep learning but ultimately Jouppi says he never excepted to return back to what is essentially a CISC device.

We are, of course, talking about Google’s Tensor Processing Unit (TPU), which has not been described in much detail or benchmarked thoroughly until this week. Today, Google released an exhaustive comparison of the TPU’s performance and efficiencies compared with Haswell CPUs and Nvidia Tesla K80 GPUs. We will cover that in more detail in a separate article so we can devote time to an in-depth exploration of just what’s inside the Google TPU to give it such a leg up on other hardware for deep learning inference. You can take a look at the full paper, which was just released, and read on for what we were able to glean from Jouppi that the paper doesn’t reveal.

Continue reading “First In-Depth Look at Google’s TPU Architecture” »

Taking a cue from the Marvel Universe, researchers report that they have developed a self-healing polymeric material with an eye toward electronics and soft robotics that can repair themselves. The material is stretchable and transparent, conducts ions to generate current and could one day help your broken smartphone go back together again.

The researchers will present their work today at the 253rd National Meeting & Exposition of the American Chemical Society (ACS).

“When I was young, my idol was Wolverine from the X-Men,” Chao Wang, Ph.D., says. “He could save the world, but only because he could heal himself. A self-healing material, when carved into two parts, can go back together like nothing has happened, just like our human skin. I’ve been researching making a self-healing lithium ion battery, so when you drop your cell phone, it could fix itself and last much longer.”

Continue reading “Materials may lead to self-healing smartphones” »

If you drop your phone and the screen shatters, you usually have two options: get it repaired or replace the phone entirely.

Chemists at the University of California, Riverside, have invented what could become a third option: a phone screen material that can heal itself.

The researchers conducted several tests on the material, including its ability to repair itself from cuts and scratches.

Continue reading “This New Smartphone Screen Material Can Repair Its Own Scratches” »

Sorry, Google and Apple: The smartphone, one day, will die. Here’s what comes after.