Lifeboat Foundation

Researchers have found a better way to reduce gender bias in natural language processing models while preserving vital information about the meanings of words, according to a recent study that could be a key step toward addressing the issue of human biases creeping into artificial intelligence.

While a computer itself is an unbiased machine, much of the data and programming that flows through computers is generated by humans. This can be a problem when conscious or unconscious human biases end up being reflected in the text samples AI models use to analyze and “understand” language.

Computers aren’t immediately able to understand text, explains Lei Ding, first author on the study and graduate student in the Department of Mathematical and Statistical Sciences. They need words to be converted into a set of numbers to understand them—a process called word embedding.

Water is the one thing all life on Earth needs, and the cycle of rain to river to ocean to rain is an essential part of what keeps our planet’s climate stable and hospitable. When scientists talk about where to search for signs of life throughout the galaxy, planets with water are always at the top of the list.

A new study published in Science suggests that many more planets may have large amounts of water than previously thought—as much as half water and half rock. The catch? All that water is probably embedded in the rock, rather than flowing as oceans or rivers on the surface.

“It was a surprise to see evidence for so many water worlds orbiting the most common type of star in the galaxy,” said Rafael Luque, first author on the new paper and a postdoctoral researcher at the University of Chicago. “It has enormous consequences for the search for habitable planets.”

Over the last decade, improvements in optical atomic clocks have repeatedly led to devices that have broken records for their precision (see Viewpoint: A Boost in Precision for Optical Atomic Clocks). To achieve even better performance, physicists must find a way to cool the atoms in these clocks to lower temperatures, which would allow them to use shallower atom traps and reduce measurement uncertainty. Tackling this challenge, Xiaogang Zhang and colleagues at the National Institute of Standards and Technology, Colorado, have cooled a gas of ytterbium atoms to a record low temperature of a few tens of nanokelvin [1]. As well as enabling the next generation of optical atomic clocks, the researchers say that their technique could be used to cool atoms in neutral-atom quantum computers.

Divalent atoms such as ytterbium are especially suited to precision metrology, as their lack of net electronic spin makes them less sensitive than other species to environmental noise. These atoms can be cooled to the necessary sub-µK temperatures in several ways, but not all techniques are compatible with the requirements of high-precision clocks. For example, evaporative cooling, in which the most energetic atoms are removed, is time-consuming and depletes the atoms. Meanwhile, resolved sideband cooling chills the motion of the atoms only along the axis of the 1D optical trap, leaving their off-axis motion unaffected.

Zhang and colleagues cool their atoms using a laser tuned to ytterbium’s so-called clock transition, whose extremely narrow linewidth means that the atom can theoretically be cooled to below 10 nK. They demonstrate that the precision of a clock employing a shallow lattice trap enabled by such a temperature would not be limited by atoms tunneling between adjacent lattice sites, potentially allowing a measurement uncertainty below 10^-19.

Magnetorotational instability—a process that might explain the dynamics of astrophysical accretion disks—has finally been observed in the laboratory.

What do black holes, forming stars, and a tank of liquid metal in Princeton, New Jersey, have in common? The first two might and the third one definitely does play host to an important process in magnetized-fluid dynamics called magnetorotational instability (MRI). MRI has been well studied theoretically and computationally, and related processes have been seen experimentally [1]. But until now, there has not been an unambiguous laboratory confirmation of its existence. Yin Wang and his colleagues at Princeton University have demonstrated MRI in an ingenious liquid-metal experiment—the culmination of more than 20 years of work [2].

The team’s discovery is significant because MRI has long been suspected of being at the heart of accretion [3]. Accretion, in which material spirals inward in a flattened disk around a black hole or a young star, is a major source of the light coming from those objects. For accretion to occur, the material in the disk must lose its angular momentum. However, angular momentum is conserved: much like the trash we generate in our daily lives, it does not cease to exist when it is not wanted. Instead, angular momentum must be passed from the inner parts of the disk to the outer parts. What drives this angular-momentum transport has long been a mystery.

On August 12, the hybrid magnet of the Steady High Magnetic Field Facility (SHMFF) in Hefei, China, generated the world’s highest steady magnetic field by a working magnet measuring 45.22 teslas (T). In comparison, Earth’s magnetic field at 0° latitude and 0° longitude only has a strength of 0.000032 teslas.

It surpassed the 45-tesla prior world record set by a hybrid magnet in 1999 at the National High Magnetic Field Laboratory of the United States.

It will be commercially available in early 2023.

London-based SolarBotanic Trees (SBT) unveiled its solar and storage system that is shaped liked a tree. The company aims to deploy its technology to charge electric vehicles (EVs) to begin with, Electrek.

With the world moving towards less-carbon emissions, there is a rush toward harnessing renewable sources of energy. Not only do these technologies need improvements in their power generation efficiencies, but they also need to be aesthetically pleasing.

Musk texted a Morgan Stanley banker, two weeks after he publicly announced his intent to buy Twitter.

The potential of World War III appears to be a reason why the world’s richest person, Elon Musk, wanted to call off his buyout offer for Twitter, Business Insider.

Wikimedia Commons.

Continue reading “Elon Musk wanted to back out of Twitter deal because of World War III. Here’s what the lawyers said in court” »

In 1998 I was exposed to the term “disruptive innovation” for the first time. I read a wonderful book, “The Innovator’s Dilemma” by Clayton Christensen, where I learned the difference between incremental innovation and disruptive innovation. He analyzed the hard drive industry and showed that while many companies were trying to increase the capacity of the drives, other companies changed the form factor and made the drives smaller. This resulted in disruptive progress in the industry. We also recently witnessed dramatic advances in artificial intelligence (AI), where in 2013/2014 AI systems started outperforming humans in image recognition.

Identifying novel targets and designing novel molecules is not the only way to innovate in the biopharmaceutical industry. Sometimes, innovation in delivery systems for well-know and established therapeutics may be just as disruptive as the new targeted medicine.

Nuro, a Softbank-backed developer of street-legal autonomous, electric delivery vehicles, has struck a long-term partnership with Uber to use its toaster-shaped micro-vans to haul food orders, groceries and other goods to customers in Silicon Valley and Houston using the Uber Eats service starting this year.

People using the Uber Eats app in Houston and Mountain View, California (where Nuro is based) will be able to order deliveries using the new autonomous service this fall, with plans to expand the program to other parts of the San Francisco Bay Area in the months ahead, the companies said.

The SoftBank-backed developer of street-legal autonomous, electric vehicles, has a long-term partnership with Uber to use its toaster-shaped micro-vans to haul food orders, groceries and other goods in Silicon Valley and Houston.

Continue reading “Uber Taps Nuro’s Street-Legal Robots For Food Deliveries” »