Lifeboat Foundation

Words categorize the semantic fields they refer to in ways that maximize communication accuracy while minimizing complexity. Recent studies have shown that human languages are optimally balanced between accuracy and complexity. For example, many languages have a word that denotes the color red, but no language has individual words to distinguish ten different shades of the color. These additional words would complicate the vocabulary and rarely would they be useful to achieve precise communication.

A study published on 23 March in the journal Proceedings of the National Academy of Sciences analyzed how artificial neural networks develop spontaneous systems to name colors. A study by Marco Baroni, ICREA research professor at the UPF Department of Translation and Language Sciences (DTCL), conducted with members of Facebook AI Research (France).

For this study, the researchers formed two artificial neural networks trained with two generic deep learning methods. As Baroni explains: “we made the networks play a color-naming game in which they had to communicate about color chips from a continuous color space. We did not limit the “language” they could use, however, when they learned to play the game successfully, we observed the color-naming terms these artificial neural networks had developed spontaneously.”

Over the past decade or so, the semimetal graphene has attracted substantial interest among electronics engineers due to its many advantageous qualities and characteristics. In fact, its high electron mobility, flexibility and stability make it particularly desirable for the development of next-generation electronics.

Despite its advantageous properties, large-area graphene has a zero bandgap (i.e., the energy range in solid materials at which no electronic states can exist). This means that electric current in graphene cannot be completely shut off. This characteristic makes it unsuitable for the development of many electronic devices.

Researchers at Tsinghua University in China recently devised a design strategy that could be used to attain a larger bandgap in graphene. This strategy, introduced in a paper published in Nature Electronics, entails the use of an electric field to control conductor-to-insulator transitions in microscale graphene.

A team of researchers from QuTech in the Netherlands reports realization of the first multi-node quantum network, connecting three quantum processors. In addition, they achieved a proof-of-principle demonstration of key quantum network protocols. Their findings mark an important milestone toward the future quantum internet and have now been published in Science.

The power of the internet is that it allows any two computers on Earth to connect. Today, researchers in many labs around the world are working toward first versions of a quantum internet—a network that can connect any two quantum devices, such as quantum computers or sensors, over large distances. Whereas today’s internet distributes information in bits that can be either 0 or 1, a future quantum internet will make use of quantum bits that can be 0 and 1 at the same time.

“A quantum internet will open up a range of novel applications, from unhackable communication and cloud computing with complete user privacy to high-precision time-keeping,” says Matteo Pompili, Ph.D. student and a member of the research team. “And like with the internet 40 years ago, there are probably many applications we cannot foresee right now.”

A new generation of miniature recording probes can track the same neurons inside tiny mouse brains over weeks—and even months.

The new tools build on the success of the original Neuropixels probes released in 2017 and currently used in more than 400 labs. Neuropixels 2.0 are much smaller—about a third the size of their predecessors. They’re designed to record the electrical activity from more individual neurons and have the unique ability to track this activity over extended time periods. That makes them especially useful for studying long-term phenomena like learning and memory in small animals such as mice, says Tim Harris, a senior fellow at HHMI’s Janelia Research Campus who led the project. Harris and his colleagues describe the advance in a paper published online April 15 in the journal Science.

Neuropixels 2.0’s advances come from several key innovations, Harris says. Janelia scientists and engineers developed new ways to process the data. Strategic changes to the layout of the probes helped make them better suited to certain tasks. And engineers at imec, the non-profit nanoelectronics center that manufactures the probes, used imec’s proprietary technology to design, develop, and fabricate the probe.

The Lightmatter photonic computer is 10 times faster than the fastest NVIDIA artificial intelligence GPU while using far less energy. And it has a runway for boosting that massive advantage by a factor of 100, according to CEO Nicholas Harris.

In the process, it may just restart a moribund Moore’s Law.

Or completely blow it up.

The lab-grown meat product ‘incorporates muscle and fat similar to its slaughtered counterpart’, the firm says.

But the latest work has divided developmental biologists. Some question the need for such experiments using closely related primates — these animals are not likely to be used as model animals in the way that mice and rodents are. Nonhuman primates are protected by stricter research ethics rules than are rodents, and they worry such work is likely to stoke public opposition.

The chimaeras lived up to 19 days — but some scientists question the need for such research.

Local health authorities in several parts of the United States are monitoring dozens of travelers for Ebola after the Centers for Disease Control (CDC) ordered airlines to collect information on people who’d been in several western African countries, including Guinea or the Democratic Republic of Congo.

Washington state is monitoring at least 23 travelers. Another 45 are being watched in Ohio. Four people are also being monitored in Oregon.

Ebola is highly contagious and causes severe illness that often leads to death. Symptoms include fever, headache, pain and unexplained bleeding or bruising.

Spot was apparently being used for reconnaissance.

Pictures of the exercises were shared on Twitter by France’s foremost military school, the École Spéciale Militaire de Saint-Cyr. It described the tests as “raising students’ awareness of the challenges of tomorrow,” which include the “robotization of the battlefield.”

A report by French newspaper Ouest-France offers more detail, saying that Spot was one of a number of robots being tested by students from France’s École Militaire Interarmes (Combined Arms School), with the intention of assessing the usefulness of robots on future battlefields.

Continue reading “The French army is testing Boston Dynamics’ robot dog Spot in combat scenarios” »