Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog – Page 4687

Improving the efficiency of algorithms for fundamental computations is a crucial task nowadays as it influences the overall pace of a large number of computations that might have a significant impact. One such simple task is matrix multiplication, which can be found in systems like neural networks and scientific computing routines. Machine learning has the potential to go beyond human intuition and beat the most exemplary human-designed algorithms currently available. However, due to the vast number of possible algorithms, this process of automated algorithm discovery is complicated. DeepMind recently made a breakthrough discovery by developing AplhaTensor, the first-ever artificial intelligence (AI) system for developing new, effective, and indubitably correct algorithms for essential operations like matrix multiplication. Their approach answers a mathematical puzzle that has been open for over 50 years: how to multiply two matrices as quickly as possible.

AlphaZero, an agent that showed superhuman performance in board games like chess, go, and shogi, is the foundation upon which AlphaTensor is built. The system expands on AlphaZero’s progression from playing traditional games to solving complex mathematical problems for the first time. The team believes this study represents an important milestone in DeepMind’s objective to improve science and use AI to solve the most fundamental problems. The research has also been published in the established Nature journal.

Matrix multiplication has numerous real-world applications despite being one of the most simple algorithms taught to students in high school. This method is utilized for many things, including processing images on smartphones, identifying verbal commands, creating graphics for video games, and much more. Developing computing hardware that multiplies matrices effectively consumes many resources; therefore, even small gains in matrix multiplication efficiency can have a significant impact. The study investigates how the automatic development of new matrix multiplication algorithms could be advanced by using contemporary AI approaches. In order to find algorithms that are more effective than the state-of-the-art for many matrix sizes, AlphaTensor further leans on human intuition. Its AI-designed algorithms outperform those created by humans, which represents a significant advancement in algorithmic discovery.

Read more | >

New to nerfstudio? Here we walk you through a step-by-step process on how to turn your favorite capture into a trendy 3D video in minutes.

Github: github.com/nerfstudio-project/nerfstudio.
Discord: discord.gg/uMbNqcraFc.
Twitter: @nerfstudioteam.

Getting started.
0:00 Hello from nerfstudio.
0:13 Preprocess your video.
0:27 Launching training and viewer.

Viewer basics.

Continue reading “Nerfstudio Viewer Tutorial” | >

❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambdalabs.com/papers.

📝 My paper “The flow from simulation to reality” with clickable citations is available here:
https://www.nature.com/articles/s41567-022-01788-5
📝 Read it for free here! https://rdcu.be/cWPfD

❤️ Watch these videos in early access on our Patreon page or join us here on YouTube:
- https://www.patreon.com/TwoMinutePapers.
- https://www.youtube.com/channel/UCbfYPyITQ-7l4upoX8nvctg/join.

🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:

Continue reading “Wow, A Simulation That Looks Like Reality! 🤯” | >

Science-fiction author Neal Stephenson predicted cryptocurrencies and virtual worlds — even coining the term metaverse — in his 1992 novel Snow Crash. Now, 30 years later, he is combining the two ideas by launching a blockchain-powered open metaverse. It is a move that sets him against big tech firms with their own metaverse plans, so why is he doing it and what even is an open metaverse?

Developers have pushed various versions of the metaverse for decades, but nothing has stuck.

Lamina1 was contacted for comment.

Neal Stephenson was contacted for comment.

Continue reading “Sci-fi author Neal Stephenson wants to build a metaverse open to all” | >

Circa 2015 face_with_colon_three

From driving water wheels to turning turbines, waterhas been used as the prime mover of machinery and the powerhouse of industry for many centuries. In ancient times, the forces of flowing water were even harnessed to power the first rudimentaryclocks. Now, engineers at Stanford University have created the world’s first water-operated computer. Using magnetized particles flowing through a micro-miniature network ofchannels, the machine runs like clockwork and is claimed to be capable ofperforming complex logical operations.

Using poppy-seed sizeddroplets of water impregnated with magnetic nanoparticles (those handy little elementsbeing used in everything from drug delivery inhumans to creating e-paper whiteboards), the new fluidic computer uses electromagnetic fields to accurately pump thesedroplets around a set of physical gates to perform logical operations. Suspendedin oil and timed to move in very specific steps, the droplets in the system cantheoretically be used to accomplish any process that a normal electroniccomputer can, albeit at considerably slower speeds.

Continue reading “Engineers create a computer with a water droplet processor” | >

Circa 2014 face_with_colon_three

A liquid hard drive containing a suspension of nanoparticles could be used to store impressive amounts of data: 1 terabyte per tablespoon.

Researchers from the University of Michigan and New York University have been simulating wet information storage techniques which uses clusters of nanoparticles suspended in liquid. These clusters of particles can store more data than conventional computer bits which have just two storage states: 0 and 1. The clusters of particles work a bit like Rubik’s Cubes to reconfigure in different ways to represent different storage states. A 12-particle memory cluster connected to a central sphere can have almost eight million unique states, which is equivalent to 2.86 bytes of data.

The system works by having nanoparticles attached to a central sphere. When the sphere is small, the outer particles trap each other into place, storing data. If the sphere is a bit larger, the particles can be reconfigured to store different information. The team created a cluster involving four particles on a central sphere — all made of polymers. By heating the liquid up, the spheres expand and the particles can rearrange themselves in predictable ways. Although the four-particle clusters have only two distinguishable configurations (i.e. like a regular bit), the plan is create clusters with many more particles.

Continue reading “Liquid hard drive could store 1TB data in a tablespoon” | >

WEIGHT IS ONE of the biggest banes for car designers and engineers. Batteries are exceedingly heavy and dense, and with the internal combustion engine rapidly pulling over for an electric future, the question of how to deal with an EV’s added battery mass is becoming all the more important.

But what if you could integrate the battery into the structure of the car so that the cells could serve the dual purpose of powering the vehicle and serving as its skeleton? That is exactly what Tesla and Chinese companies such as BYD and CATL are working on. The new structural designs coming out of these companies stand to not only change the way EVs are produced but increase vehicle ranges while decreasing manufacturing costs.

Auto companies are designing ways to build a car’s fuel cells into its frame, making electric rides cheaper, roomier, and able to hit ranges of 620 miles.

Continue reading “A New Approach to Car Batteries Is About to Transform EVs” | >

Prof. Ehud Pines (pictured above) is an iconoclast. What else can you call a scientist who spent 17 years doggedly pursuing the solution to an over 200-year-old chemistry problem that he felt never received a satisfying answer using methods no other scientist thought could lead to the truth? Now, he is vindicated as the prestigious Angewandte Chemie journal published a cover article detailing how his experiment was replicated by another research group while being x-rayed to reveal the solution Prof. Pines has argued for all along.

The question at hand is: How does a proton move through water? In 1,806, Theodor Grotthuss proposed his theory, which became known as the Grotthuss Mechanism. Over the years, many others attempted an updated solution realizing that strictly speaking, Grotthuss was incorrect, but it remained the standard textbook answer. Until now.

Prof. Ehud Pines suggested, based on his experimental studies at Ben-Gurion University of the Negev in the Department of Chemistry, together with his PhD student Eve Kozari, and theoretical studies by Prof. Benjamin Fingerhut on the structure of Prof. Pines’ protonated water clusters, that the proton moves through water in trains of three water molecules. The proton train “builds the tracks” underneath them for their movement and then disassembles the tracks and rebuilds them in front of them to keep going. It’s a loop of disappearing and reappearing tracks that continues endlessly. Similar ideas were put forward by a number of scientists in the past, however, according to Prof. Pines, they were not assigned to the correct molecular structure of the hydrated proton which by its unique trimeric structural properties leads to promoting the Grotthuss mechanism.

Read more | >