Lifeboat Foundation

We search for the signature of parity-violating physics in the cosmic microwave background, called cosmic birefringence, using the Planck data release 4. We initially find a birefringence angle of β=0.30±0.11 (68% C.L.) for nearly full-sky data. The values of β decrease as we enlarge the Galactic mask, which can be interpreted as the effect of polarized foreground emission. Two independent ways to model this effect are used to mitigate the systematic impact on β for differen… See more.

We search for the signature of parity-violating physics in the cosmic.

Microwave background, called cosmic birefringence, using the Planck data.
release 4. We initially find a birefringence angle of $\beta=0.30\pm0.11$ (68%

Continue reading “Cosmic Birefringence from Planck Data Release 4” »

What’s next? Human brain-scale AI.

Funded by the Slovakian government using funds allocated by the EU, the I4DI consortium is behind the initiative to build a 64 AI exaflop machine (that’s 64 billion, billion AI operations per second) on our platform by the end of 2022. This will enable Slovakia and the EU to deliver for the first time in the history of humanity a human brain-scale AI supercomputer. Meanwhile, almost a dozen other countries are watching this project closely, with interest in replicating this supercomputer in their own countries.

There are multiple approaches to achieve human brain-like AI. These include machine learning, spiking neural networks like SpiNNaker, neuromorphic computing, bio AI, explainable AI and general AI. Multiple AI approaches require universal supercomputers with universal processors for humanity to deliver human brain-scale AI.

Advances in the AI realm are constantly coming out, but they tend to be limited to a single domain: For instance, a cool new method for producing synthetic speech isn’t also a way to recognize expressions on human faces. Meta (AKA Facebook) researchers are working on something a little more versatile: an AI that can learn capably on its own whether it does so in spoken, written or visual materials.

The traditional way of training an AI model to correctly interpret something is to give it lots and lots (like millions) of labeled examples. A picture of a cat with the cat part labeled, a conversation with the speakers and words transcribed, etc. But that approach is no longer in vogue as researchers found that it was no longer feasible to manually create databases of the sizes needed to train next-gen AIs. Who wants to label 50 million cat pictures? Okay, a few people probably — but who wants to label 50 million pictures of common fruits and vegetables?

Currently some of the most promising AI systems are what are called self-supervised: models that can work from large quantities of unlabeled data, like books or video of people interacting, and build their own structured understanding of what the rules are of the system. For instance, by reading a thousand books it will learn the relative positions of words and ideas about grammatical structure without anyone telling it what objects or articles or commas are — it got it by drawing inferences from lots of examples.

Many people think that mathematics is a human invention. To this way of thinking, mathematics is like a language: it may describe real things in the world, but it doesn’t “exist” outside the minds of the people who use it.

But the Pythagorean school of thought in ancient Greece held a different view. Its proponents believed reality is fundamentally mathematical. More than 2,000 years later, philosophers and physicists are starting to take this idea seriously.

Continue reading “Humans Didn’t Invent Mathematics, It’s What the World Is Made Of” »

Circa 2016 o.o!

A ruby that formed in Earth’s crust 2.5 billion years ago encases evidence for early life, wriggling around in the planet’s mud.

Keeping up with the first law of robotics: a new photonic effect for accelerated drug discovery. Physicists at the University of Bath and University of Michigan demonstrate a new photonic effect in semiconducting nanohelices. A new photonic effect in semiconducting helical particles with nanos.

California has more rooftops with solar panels than any other state and continues to be a leader in new installations. It is also first in terms of the percentage of the state’s electricity coming from solar, and third for solar power capacity per capita. However, former California governor Arnold Schwarzenegger has expressed concerns that California.

California has more rooftops with solar panels than any other state and continues to be a leader in new installations. It is also first in terms of the percentage of the state’s electricity coming from solar, and third for solar power capacity per capita. However, former California governor Arnold Schwarzenegger has expressed concerns that California solar — once the model for other US states — is on a precipice. In an opinion piece for the New York Times this week, Schwarzenegger has unpacked a new California Public Utilities Commission proposal which, if approved, would discourage progress being made in the transition to clean energy and grid resilience.

What’s the problem, then? The California Public Utilities Commission is threatening solar progress. But this “hard-earned and vitally important accomplishment” may succumb as the Commission considers a plan that has the potential to make California solar too costly for its citizens.

ASML President and CTO Martin van den Brink said:

“Intel’s vision and early commitment to ASML’s High-NA EUV technology is proof of its relentless pursuit of Moore’s Law. Compared to the current EUV systems, our innovative extended EUV roadmap delivers continued lithographic improvements at reduced complexity, cost, cycle time and energy that the chip industry needs to drive affordable scaling well into the next decade.”

Intel plans to start high-volume manufacturing (HVM) in 2025, which is also when the company will be using its 18A (1.8nm) fabrication technology. To do so, Intel has been experimenting for quite a while when it first obtained ASML’s Twinscan EXE:5000, which was the industry’s first EUV scanner with a 0.55 numerical aperture. Today, the company ordered ASML’s next-generation High-NA tool, the Twinscan EXE:5200.

High-tech missiles, sensors, aircraft, ships, and artillery all rely on atomic clocks on GPS satellites for nanosecond timing accuracy. A timing error of just a few billionths of a second can translate to positioning being off by a meter or more. If GPS were jammed by an adversary, time synchronization would rapidly deteriorate and threaten military operations.

To address this scenario, DARPA has announced the Robust Optical Clock Network (ROCkN) program, which aims to create optical atomic clocks with low size, weight, and power (SWaP) that yield timing accuracy and holdover better than GPS atomic clocks and can be used outside a laboratory. ROCkN will leverage DARPA-funded research over the past couple decades that has led to lab demonstration of the world’s most precise optical atomic clocks. ROCkN clocks will not be as precise as the best lab optical clocks, but they will surpass current state-of-the-art atomic clocks in both precision and holdover while maintaining low SWaP in a robust package. https://www.darpa.mil/news-events/2022-01-20