Are you ready for artificial intelligence?
Curt’s “String Theory Iceberg”: https://youtu.be/X4PdPnQuwjYMain episode with Bach and Goertzel (October 2023): https://youtu.be/xw7omaQ8SgA?list=PLZ7ikzmc6zlN6E8KrxcYCWQIHg2tfkqvR Consider signing up for TOEmail at https://www.curtjaimungal.org
Scientists at Argonne National Laboratory have developed a new material that could replace expensive nickel alloys in nuclear reactors.
A research group led by Prof. Liang Xu from the Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science of Chinese Academy of Sciences, propose a coupling analysis model revealing the flow characteristics and control laws of a multi-needle Electrohydrodynamics (EHD) pump.
An experiment aimed at learning more about how plants grow in space will be aboard a National Aeronautics and Space Administration launch in early August from the Cape Canaveral Space Force Station in Florida.
The “super premium” segment here implies a driving range of around 600 miles per charge. In addition, Samsung will be introducing high-nickel NCS products for the premium segment.
Samsung’s oxide solid-state battery technology boasts an energy density of 500 Wh/kg, nearly double the 270 Wh/kg density of mainstream EV batteries.
Continue reading “Samsung’s 20-year-life EV battery runs 600 miles on 9-minute charge” »
This could be the road to quantum computation.
“In contrast, solid-state emitters embedded in a photonic circuit are hardly ‘the same’ due to slightly different surroundings influencing each emitter. It is much harder for many solid-state emitters to build up phase coherence and collectively interact with photons like cold atoms. We could use cold atoms trapped on the circuit to study new collective effects,” Hung continues.
The platform demonstrated in this research could provide a photonic link for future distributed quantum computing based on neutral atoms. It could also serve as a new experimental platform for studying collective light-matter interactions and for synthesizing quantum degenerate trapped gases or ultracold molecules.
Continue reading “Researchers trap atoms, force them to serve as photonic transistors” »
It looks like NASA officials might be seeing the writing on the wall for the very troubled Boeing Starliner, which has marooned two astronauts up in space for almost two months due to technical issues.
An unnamed “informed” source told Ars Technica that there’s a greater than 50 percent probability that the stranded astronauts will end up leaving the International Space Station on a SpaceX Dragon capsule, with another unnamed person telling the news outlet that the scenario is highly likely.
NASA officials are more cagey about what’s happening on the record, a marked contrast from previous weeks when they expressed confidence in the Starliner’s ability to safely bring back the astronauts.
One of the significant challenges in AI research is the computational inefficiency in processing visual tokens in Vision Transformer (ViT) and Video Vision Transformer (ViViT) models. These models process all tokens with equal emphasis, overlooking the inherent redundancy in visual data, which results in high computational costs. Addressing this challenge is crucial for the deployment of AI models in real-world applications where computational resources are limited and real-time processing is essential.
Current methods like ViTs and Mixture of Experts (MoEs) models have been effective in processing large-scale visual data but come with significant limitations. ViTs treat all tokens equally, leading to unnecessary computations. MoEs improve scalability by conditionally activating parts of the network, thus maintaining inference-time costs. However, they introduce a larger parameter footprint and do not reduce computational costs without skipping tokens entirely. Additionally, these models often use experts with uniform computational capacities, limiting their ability to dynamically allocate resources based on token importance.
In the next 20 years, AI will eat software.
“We made it possible for the computer to write software by itself.”
—NVIDIA CEO Jensen Huang on the future of AI
In 2021, a team led by MIT physicists reported creating a new ultrathin ferroelectric material, or one where positive and negative charges separate into different layers. At the time, they noted the material’s potential for applications in computer memory and much more. Now the same core team and colleagues—including two from the lab next door—have built a transistor with that material and shown that its properties are so useful that it could change the world of electronics.
Although the team’s results are based on a single transistor in the lab, “in several aspects its properties already meet or exceed industry standards” for the ferroelectric transistors produced today, says Pablo Jarillo-Herrero, the Cecil and Ida Green Professor of Physics, who led the work with professor of physics Raymond Ashoori. Both are also affiliated with the Materials Research Laboratory.
“In my lab we primarily do fundamental physics. This is one of the first, and perhaps most dramatic, examples of how very basic science has led to something that could have a major impact on applications,” Jarillo-Herrero says.
