Lifeboat Foundation

Start listening with a 30-day Audible trial and your first audiobook is free. Visit http://www.audible.com/isaac or text “isaac” to 500–500.
The Silurian Hypothesis contemplates how long the ruins of a civilization would be detectable, on Earth or even other worlds, and if we could ever know if a world had once been inhabited by a technological civilization.

Visit our Website: http://www.isaacarthur.net.
Support us on Patreon: https://www.patreon.com/IsaacArthur.
Support us on Subscribestar: https://www.subscribestar.com/isaac-arthur.
Facebook Group: https://www.facebook.com/groups/1583992725237264/
Reddit: https://www.reddit.com/r/IsaacArthur/
Twitter: https://twitter.com/Isaac_A_Arthur on Twitter and RT our future content.
SFIA Discord Server: https://discord.gg/53GAShE

Continue reading “The Silurian Hypothesis: What Traces Of Humanity Will Be Left 50 Million Years From Now?” »

A team of researchers affiliated with multiple institutions in the U.S., including Google Quantum AI, and a colleague in Australia, has developed a theory suggesting that quantum computers should be exponentially faster on some learning tasks than classical machines. In their paper published in the journal Science, the group describes their theory and results when tested on Google’s Sycamore quantum computer. Vedran Dunjko with Leiden University City has published a Perspective piece in the same journal issue outlining the idea behind combining quantum computing with machine learning to provide a new level of computer-based learning systems.

Machine learning is a system by which computers trained with datasets make informed guesses about new data. And quantum computing involves using sub-atomic particles to represent qubits as a means for conducting applications many times faster than is possible with classical computers. In this new effort, the researchers considered the idea of running machine-learning applications on quantum computers, possibly making them better at learning, and thus more useful.

To find out if the idea might be possible, and more importantly, if the results would be better than those achieved on classical computers, the researchers posed the problem in a novel way—they devised a machine learning task that would learn via experiments repeated many times over. They then developed theories describing how a quantum system could be used to conduct such experiments and to learn from them. They found that they were able to prove that a quantum computer could do it, and that it could do it much better than a classical system. In fact, they found a reduction in the required number of experiments needed to learn a concept to be four orders of magnitude lower than for classical systems. The researchers then built such a system and tested it on Google’s Sycamore quantum computer and confirmed their theory.

Nuclear fusion is a widely studied process through which atomic nuclei of a low atomic number fuse together to form a heavier nucleus, while releasing a large amount of energy. Nuclear fusion reactions can be produced using a method known as inertial confinement fusion, which entails the use of powerful lasers to implode a fuel capsule and produce plasma.

Researchers at Massachusetts Institute of Technology (MIT), University of Delaware, University of Rochester, the Lawrence Livermore National Laboratory, Imperial College London, and University of Rome La Sapienza have recently showed what happens to this implosion when one applies a strong magnetic field to the fuel capsule used for inertial confinement fusion. Their paper, published in Physical Review Letters, demonstrates that strong magnetic fields flatten the shape of inertial fusion implosions.

“In inertial confinement fusion, a millimeter-size spherical capsule is imploded using high-power lasers for nuclear fusion,” Arijit Bose, one of the researchers who carried out the study, told Phys.org. “Applying a magnetic field to the implosions can strap the charged plasma particles to the B-field and improve their chances of fusion. However, since magnetic field can restrict plasma particle motion only in the direction across the field lines and not in the direction along the applied field lines, this can introduce differences between the two directions that affect the implosion shape.”

USC scientists have found evidence that the Earth’s inner core oscillates, contradicting previously accepted models that suggested it consistently rotates at a faster rate than the planet’s surface.

Their study, published today in Science Advances, shows that the inner core changed direction in the six-year period from 1969–74, according to the analysis of seismic data. The scientists say their model of inner core movement also explains the variation in the length of day, which has been shown to oscillate persistently for the past several decades.

“From our findings, we can see the Earth’s surface shifts compared to its inner core, as people have asserted for 20 years,” said John E. Vidale, co-author of the study and Dean’s Professor of Earth Sciences at USC Dornsife College of Letters, Arts and Sciences. “However, our latest observations show that the inner core spun slightly slower from 1969–71 and then moved the other direction from 1971–74. We also note that the length of day grew and shrank as would be predicted.

Researchers at Duke University and the University of Maryland have used the frequency of measurements on a quantum computer to get a glimpse into the quantum phenomena of phase changes—something analogous to water turning to steam.

By measuring the number of operations that can be implemented on a quantum computing system without triggering the collapse of its quantum state, the researchers gained insight into how other systems—both natural and computational—meet their tipping points between phases. The results also provide guidance for computer scientists working to implement quantum error correction that will eventually enable quantum computers to achieve their full potential.

The results appeared online June 3 in the journal Nature Physics.

A major campaign of domino-toppling simulations yields new insights into the effects of friction.

Despite the apparent simplicity of toppling dominoes, physicists still don’t have a complete model of the phenomenon. But new numerical simulations get a step closer by untangling the influence of two types of friction—one between neighboring dominoes and the other between each domino and the surface beneath it [1]. The researchers found that, in some cases, these two friction coefficients play competing roles in determining the speed of the domino cascade. They also found that one of the coefficients behaves similar to friction in granular systems such as piles of sand or pharmaceutical pills, suggesting that the domino simulations may provide insights into other situations where friction is important.

Continue reading “Friction Is Key in Domino Physics” »

View insights.

0 post reach.

Noise in an electronic circuit is a nuisance that can scramble information or reduce a detector’s sensitivity. But noise also offers a way to learn about the microscopic quantum mechanisms at play in a material or device. By measuring a circuit’s “shot noise,” a form of white noise, researchers have previously shed light on conduction in quantum Hall and spintronic systems, for instance. Now, a collaboration led by Oren Tal at the Weizmann Institute of Science, Israel, and by Dvira Segal at the University of Toronto, Canada, has shown that an easier-to-measure form of noise, called “flicker noise,” can also be a powerful probe of quantum effects [1].

Continue reading “Pink Noise as a Probe of Quantum Transport” »

Researchers have cooled indium atoms to a temperature close to 1 mK, making indium the first group-III atom to be made ultracold.

At temperatures near to absolute zero, atoms move slower than a three-toed sloth, allowing physicists to gain unprecedented experimental control over these systems. New phases of matter can form when atoms become ultracold and quirky quantum properties can emerge, yet much of the periodic table remains unexplored in the ultracold regime. Now, Travis Nicholson of the National University of Singapore and colleagues have successfully cooled indium to close to 1 mK [1]. Indium is the first “main group-III” atom—a specific group of transition metals on the periodic table—to be cooled to such a low temperature. The demonstration opens the door to studying systems with properties previously unexplored by ultracold physicists.

For their experiments, Nicholson and colleagues used a magneto-optical trap—a standard tool for trapping and cooling atoms. But because this was the first attempt at making indium atoms ultracold, the team had to make their own version of the apparatus rather than using one designed to cool other atoms. “The systems used for this research are highly customized to specific atoms,” Nicholson says. So every part of the setup from designing the laser systems to picking the screws had to be “hashed out by us.” With their custom setup, the group loaded 500,000,000 indium atoms into the trap using a laser beam and then cooled them.

Abstract: Superintelligence, the next phase beyond today’s narrow AI and tomorrow’s AGI, almost intrinsically evades our attempts at detailed comprehension. Yet very different perspectives on superintelligence exist today and have concrete influence on thinking about matters ranging from AGI architectures to technology regulation.
One paradigm considers superintelligences as resembling modern deep reinforcement learning systems, obsessively concerned with optimizing particular goal functions. Another considers superintelligences as open-ended, complex evolving systems, ongoingly balancing drives.
toward individuation and radical self-transcendence in a paraconsistent way. In this talk I will argue that the open-ended conception of superintelligence is both more desirable and more realistic, and will discuss how concrete work being done today on projects like OpenCog Hyperon, SingularityNET and Hypercycle potentially paves the way for a path through beneficial decentralized integrative AGI and on to open-ended superintelligence and ultimately the Singularity.

Bio: In May 2007, Goertzel spoke at a Google tech talk about his approach to creating artificial general intelligence. He defines intelligence as the ability to detect patterns in the world and in the agent itself, measurable in terms of emergent behavior of “achieving complex goals in complex environments”. A “baby-like” artificial intelligence is initialized, then trained as an agent in a simulated or virtual world such as Second Life to produce a more powerful intelligence. Knowledge is represented in a network whose nodes and links carry probabilistic truth values as well as “attention values”, with the attention values resembling the weights in a neural network. Several algorithms operate on this network, the central one being a combination of a probabilistic inference engine and a custom version of evolutionary programming.

Continue reading “Ben Goertzel — Open Ended vs Closed Minded Conceptions of Superintelligence” »