Monthly Archives: June 2022

NEW YORK, June 03, 2022 (GLOBE NEWSWIRE) -- Bragar Eagel & Squire, P.C., a nationally recognized shareholder rights law firm, is investigating potential claims against Medallion Financial Corp. (NASDAQ: MFIN), RBB Bancorp (NASDAQ: RBB), Dentsply Sirona, Inc. (NASDAQ: XRAY), and Singularity Future Technology, Inc. (NASDAQ: SGLY). Our investigations concern whether these companies have violated the federal securities laws and/or engaged in other unlawful business practices. Additional information about each case can be found at the link provided.

Medallion Financial Corp. (NASDAQ: MFIN)

On December 29, 2021, the SEC charged Medallion and its President and Chief Operating Officer, Andrew Murstein, with illegally engaging in two schemes in an effort to reverse the companys plummeting stock price. Specifically, the two had engaged in illegal touting by paying Ichabods Cranium and others to place positive stories about the company on various websites, including Huffington Post, Seeking Alpha, and TheStreet.com.

On this news, Medallions stock fell up to 27% during intraday trading on December 29, 2021, thereby injuring investors.

For more information on the Medallion investigation go to: https://bespc.com/cases/MFIN

RBB Bancorp (NASDAQ: RBB)

On February 18, 2022, RBB Bancorp announced the abrupt departure of Tammy Song, the EVP and Chief Lending Officer of RBB Bancorps wholly owned subsidiary Royal Business Bank.

Four days later, on February 22, 2022, RBB Bancorp announced its President and CEO (Alan Thian) would take a leave of absence, effective immediately, pending an internal investigation being conducted by a special committee of the Companys board of directors.

On this news, RBB Bancorps stock price declined by $2.69 per share, or approximately 10.45%, from $25.75 to $23.06 over two trading days.

For more information on the RBB Bancorp investigation go to: https://bespc.com/cases/RBB

Dentsply Sirona, Inc. (NASDAQ: XRAY)

On April 19, 2022, the Company issued a press release announcing the termination of Chief Executive Officer, Don Casey, effective immediately, and that Casey will also cease to serve as a member of the Companys Board.

Following this news, shares of Dentsply Sirona dropped sharply by $6.52 per share, over 13%, to close at $42.20 per share on April 19, 2022.

For more information on the Dentsply Sirona investigation go to: https://bespc.com/cases/XRAY

Singularity Future Technology, Inc. (NASDAQ: SGLY)

On May 5, 2022, Hindenburg Research (Hindenburg) published a report entitled Singularity Future Technology: This Nasdaq-Listed Companys CEO Is a fugitive, on the Run for Allegedly Operating a Massive Ponzi Scheme. The Hindenburg report alleged, among other things, that singularitys CEO, Yang Jie, is a fugitive on the run from Chinese authorities for running an alleged $300 million Ponzi scheme that lured in over 20,000 victims and fled to the U.S. while at least 28 other individuals involved in the case were sentenced to prison terms ranging from 6 months to 15 years. The Hindenburg report further alleged that Singularitys massive [cryptocurrency] mining rig deal appears to be a brazen undisclosed related party deal and that [w]e see little evidence that Singularitys proprietary crypto mining rigs ever existed in the first place. The photos and descriptions of Singularitys miners match precisely with another brand called KOI Miner.

On this news, Singularitys stock price fell $1.95 per share, or 28.89%, to close at $4.80 per share on May 5, 2022.

For more information on the Singularity Future investigation go to: https://bespc.com/cases/SGLY

About Bragar Eagel & Squire, P.C.:

Bragar Eagel & Squire, P.C. is a nationally recognized law firm with offices in New York, California, and South Carolina. The firm represents individual and institutional investors in commercial, securities, derivative, and other complex litigation in state and federal courts across the country. For more information about the firm, please visit http://www.bespc.com. Attorney advertising. Prior results do not guarantee similar outcomes.

Contact Information:

Bragar Eagel & Squire, P.C.Brandon Walker, Esq. Melissa Fortunato, Esq.(212) 355-4648investigations@bespc.comwww.bespc.com

Follow this link:

Bragar Eagel & Squire, PC Is Investigating Medallion, RBB, Dentsply Sirona, and Singularity Future and Encourages Investors to Contact the Firm -...

Today, Oak Ridge National Laboratorys Frontier supercomputer was crowned fastest on the planet in the semiannual Top500 list. Frontier more than doubled the speed of the last titleholder, Japans Fugaku supercomputer, and is the first to officially clock speeds over a quintillion calculations a seconda milestone computing has pursued for 14 years.

Thats a big number. So before we go on, its worth putting into more human terms.

Imagine giving all 7.9 billion people on the planet a pencil and a list of simple arithmetic or multiplication problems. Now, ask everyone to solve one problem per second for four and half years. By marshaling the math skills of the Earths population for a half-decade, youve now solved over a quintillion problems.

Frontier can do the same work in a second, and keep it up indefinitely. A thousand years worth of arithmetic by everyone on Earth would take Frontier just a little under four minutes.

This blistering performance kicks off a new era known as exascale computing.

The number of floating-point operations, or simple mathematical problems, a computer solves per second is denoted FLOP/s or colloquially flops. Progress is tracked in multiples of a thousand: A thousand flops equals a kiloflop, a million flops equals a megaflop, and so on.

The ASCI Red supercomputer was the first to record speeds of a trillion flops, or a teraflop, in 1997. (Notably, an Xbox Series X game console now packs 12 teraflops.) Roadrunner first broke the petaflop barrier, a quadrillion flops, in 2008. Since then, the fastest computers have been measured in petaflops. Frontier is the first to officially notch speeds over an exaflop1.102 exaflops, to be exactor 1,000 times faster than Roadrunner.

Its true todays supercomputers are far faster than older machines, but they still take up whole rooms, with rows of cabinets bristling with wires and chips. Frontier, in particular, is a liquid-cooled system by HPE Cray running 8.73 million AMD processing cores. In addition to being the fastest in the world, its also the second most efficientoutdone only by a test system made up of one of its cabinetswith a rating of 52.23 gigaflops/watt.

Most supercomputers are funded, built, and operated by government agencies. Theyre used by scientists to model physical systems, like the climate or structure of the universe, but also by the military for nuclear weapons research.

Supercomputers are now tailor-made to run the latest algorithms in artificial intelligence too. Indeed, a few years ago, Top500 added a new lower precision benchmark to measure supercomputing speed on AI applications. By that mark, Fugaku eclipsed an exaflop way back in 2020. The Fugaku system set the most recent record for machine learning at 2 exaflops. Frontier smashed that record with AI speeds of 6.86 exaflops.

As very large machine learning algorithms have emerged in recent years, private companies have begun to build their own machines alongside governments. Microsoft and OpenAI made headlines in 2020 with a machine they claimed was fifth fastest in the world. In January, Meta said its upcoming RSC supercomputer would be fastest at AI in the world at 5 exaflops. (It appears theyll now need a few more chips to match Frontier.)

Frontier and other private supercomputers will allow machine learning algorithms to further push the limits. Todays most advanced algorithms boast hundreds of billions of parametersor internal connectionsbut upcoming algorithms will likely grow into the trillions.

So, exascale supercomputers will allow researchers to advance technology and do new cutting-edge science that was once impractical on slower machines.

When exactly supercomputing first broke the exaflop barrier partly depends on how you define it and whats been measured.

Folding@Home, which is a distributed system made up of a motley crew of volunteer laptops, broke an exaflop at the beginning of the pandemic. But according to Top500 cofounder Jack Dongarra, Folding@Home is a specialized system thats embarrassingly parallel and only works on problems with pieces that can be solved totally independently.

More relevantly, rumors were flying last year that China had as many as two exascale supercomputers operating in secret. Researchers published some details on the machines in papers late last year, but they have yet to be officially benchmarked by Top500. In an IEEE Spectrum interview last December, Dongarra speculated that if exascale machines exist in China, the government may be trying not to shine a spotlight on them to avoid stirring up geopolitical tensions that could drive the US to restrict key technology exports.

So, its possible China beat the US to the exascale punch, but going by the Top500, a benchmark the supercomputing fields used to determine top dog since the early 1990s, Frontier still gets the official nod.

It took about 12 years to go from terascale to petascale and another 14 to reach exascale. The next big leap forward may well take as long or longer. The computing industry continues to make steady progress on chips, but the pace has slowed and each step has become more costly. Moores Law isnt dead, but its not as steady as it used to be.

For supercomputers, the challenge goes beyond raw computing power. It might seem that you should be able to scale any system to hit whatever benchmark you like: Just make it bigger. But scale requires efficiency too, or energy requirements spiral out of control. Its also harder to write software to solve problems in parallel across ever-bigger systems.

The next 1,000-fold leap, known as zettascale, will require innovations in chips, the systems connecting them into supercomputers, and the software running on them. A team of Chinese researchers predicted wed hit zettascale computing in 2035. But of course, no one really knows for sure. Exascale, predicted to arrive by 2018 or 2020, made the scene a few years behind schedule.

Whats more certain is the hunger for greater computing power isnt likely to dwindle. Consumer applications, like self-driving cars and mixed reality, and research applications, like modeling and artificial intelligence, will require faster, more efficient computers. If necessity is the mother of invention, you can expect ever-faster computers for a while yet.

Image Credit: Oak Ridge National Laboratory (ORNL)

Original post:

Wickedly Fast Frontier Supercomputer Officially Ushers in the Next Era of Computing - Singularity Hub

Memories are like scenes in a movie, and the brain is an excellent video editor.

Take a normal day as an example. A morning routineshower, coffee, checking emailsis seamlessly cut into one continuous scene. Other goings-on throughout the day become separate cuts, so when recalling one memorya fabulous bagel sandwich deli, for examplethe memory pops into the mind on its own. Yet ultimately we still retain a continuous narrative of our lives that shapes who we are, without memories blending into each other in an incomprehensible plot.

How does the brain do that?

A new study in Nature points to a surprising cluea protein called CCR5. The name may sound familiar. Its well known as an entryway for HIV infection. Its also the gene that skyrocketed to fame in 2018, when a rogue scientist used CRISPR-Cas9 to engineer the worlds first gene-edited babies, triggering a global backlash and landing him in prison.

Part of the worry for the babies is that CCR5 is a multitasker. In the brain, for example, it dwells in cells at high levels in the hippocampus, a region critical for memory. CCR5 has previously been linked to memory functions, prompting questions on how the edits could alter the babies cognition down the line.

The new study offers additional clues. In mice, CCR5 acts as the scissor tool in video editing. As the brain continuously processes new experiences into memories, high levels of CCR5 essentially snip the timeline into distinct episodes. While normally helpful, CCR5 levels rise during aging and often become overzealous. The result is memory lossa mind that struggles to link memories into comprehensive events.

Our memories are a huge part of who we are, said study lead author Dr. Alcino Silva at the David Geffen School of Medicine at UCLA. The ability to link related experiences teaches us how to stay safe and operate successfully in the world.

The good news? Using an FDA-approved drug for HIV treatment, the team restored an aging minds ability to link memories in mice, a result that could have substantial clinical implications, the team said.

Memories may seem like amorphous wisps inside a mind, but they have strong neurobiological underpinnings.

When we encounter new experiences, a select group of neurons inside a part of the hippocampus encodes these data. How these neurons are selected is still relatively mysterious, but scientists have engineered ways to prime certain neurons to participate in the memory so that theyre more active and likely to participate.

Once a memory is stored, this network of neurons is called an engram. The process is very loosely similar to a computer allocating memory storage to save a new document. Activating any component of the engram triggers the whole network to activate, which retrieves the memory.

This process relies on a dance of a myriad of proteins. Neurons are like hotels with multiple protein locksreceptorsand keys, called ligands. The locks are generally dotted on the surface of the cell. Each requires the right key to unlock it. Unlocking triggers a cascade of molecule signaling inside cells, which eventually changes how a neuron behaves; it might get more excitable, or more inhibited and less inclined to network with others. Less networking means less memory encoding, and potentially less memory linking.

How does the brain know that two events are occurring closely in time? Addressing this question requires an understanding of how time is encoded in memorya major unknown in memory research, said Andrea Teceros and Dr. Priya Rajasethupathy at the Rockefeller University, who were not involved in the study.

In the new study, the team honed in on CCR5 as a key protein to delink a memory stream. Although mostly known as part of the immune system, CCR5 is a receptor highly expressed in the hippocampus, and previous studies suggested a potential role in memory.

The team began with a popular setup for memory tests for mice. They first placed the mice into one cage to explore, and five hours later, placed them into a different cage with wildly different decorations. Here the mice received a quick and mild electrical zap, enough to startle them and make them freeze in fear. This encoded two different memories. Two days later, when placed back into the first benign cage, the mice also frozeshowing that theyve linked the first cage and the second, where they actually received the shock, into one memory.

The time gap was key. When the team extended the five-hour gap between the two cages to 24 hours, the mice could no longer link the memories.

The switch turned out to be CCR5. The protein levels briefly shot up 12 hours after the initial encoding, and tanked soon aftera trajectory that parallels the time course for memory linking. Genetically deleting or inhibiting CCR5 prolonged the memory linking window so that the mice could still hook up memories of the two cages up to seven days apart. In contrast, increasing CCR5 further dampened the mices ability to link those memories.

Digging deeper, CCR5 seems to directly tinker with neural activity in the hippocampusthe memory ledgerto dampen its activity. Two memories close in time often share overlapping engrams. Here, mice without CCR5 had greater overlaps between memories of the two cages and higher memory linking compared to normal counterparts.

Like most immune molecules, CCR5 levels rise with age, suggesting they may increasingly hack apart memories. This led the team to wonder: can we inhibit the protein to boost memory function with age?

They tested the theory in middle-aged mice with an FDA-approved drug for HIV, maraviroc, that inhibits the proteins functions. A single infusion directly into the memory center hippocampus improved the aging animals ability to link memories, spread five hours apart, in a subsequent test. Similarly, middle-aged mice genetically lacking CCR5 also linked up memories better than their normal peers.

When we gave maraviroc to older mice, the drug duplicated the effect of genetically deleting CCR5 from their DNAthe older animals were able to link memories again, said Silva.

To Terceros and Rajasthupathy, the results could help with memory loss. Because memory deficits in aging and in Alzheimers disease might be dominated by deficits in retrieval (and thus memory linking) rather than memory storage, which involves separate brain circuits, these results could have clinical implications.

The study is one of the first to interrogate the brains clock for memory linking and segregation. It opens up a world of new questions. How does memory linking affect learning? Can it help time-stamp memories during storage in the mind? As an immune molecule, how does CCR5 further influence memory and other cognitive functions? And what, if anything, happens when human embryos are shoddily deprived of the molecule, as in the case of the CRISPR babies?

For now, the team is mostly eyeing CCR5s therapeutic potential. Our next step will be to organize a clinical trial to test maravirocs influence on early memory loss with the goal of early intervention, said Silva. Once we fully understand how memory declines, well possess the potential to slow down the process.

Image Credit: geralt / 24463 images

Read the rest here:

A Critical Immune Protein Helps the Brain Link Memories, and Could Combat Aging - Singularity Hub

Humans are unrivaled in the area of cognition. After all, no other species has sent probes to other planets, produced lifesaving vaccines, or created poetry. How information is processed in the human brain to make this possible is a question that has drawn endless fascination, yet no definitive answers.

Our understanding of brain function has changed over the years. But current theoretical models describe the brain as a distributed information-processing system. This means it has distinct components that are tightly networked through the brains wiring. To interact with each other, regions exchange information though a system of input and output signals.

However, this is only a small part of a more complex picture. In a study published last week in Nature Neuroscience, using evidence from different species and multiple neuroscientific disciplines, we show that there isnt just one type of information processing in the brain. How information is processed also differs between humans and other primates, which may explain why our species cognitive abilities are so superior.

We borrowed concepts from whats known as the mathematical framework of information theorythe study of measuring, storing, and communicating digital information which is crucial to technology such as the internet and artificial intelligenceto track how the brain processes information. We found that different brain regions in fact use different strategies to interact with each other.

Some brain regions exchange information with others in a very stereotypical way, using input and output. This ensures that signals get across in a reproducible and dependable manner. This is the case for areas that are specialized for sensory and motor functions (such as processing sound, visual, and movement information).

Take the eyes, for example, which send signals to the back of the brain for processing. The majority of information that is sent is duplicate, being provided by each eye. Half of this information, in other words, is not needed. So we call this type of input-output information processing redundant.

But the redundancy provides robustness and reliability; it is what enables us to still see with only one eye. This capability is essential for survival. In fact, it is so crucial that the connections between these brain regions are anatomically hard-wired in the brain, a bit like a telephone landline.

However, not all information provided by the eyes is redundant. Combining information from both eyes ultimately enables the brain to process depth and distance between objects. This is the basis for many kinds of 3D glasses at the cinema.

This is an example of a fundamentally different way of processing information, in a way that is greater than the sum of its parts. We call this type of information processingwhen complex signals from across different brain networks are integratedsynergistic.

Synergistic processing is most prevalent in brain regions that support a wide range of more complex cognitive functions, such as attention, learning, working memory, and social and numerical cognition. It is not hardwired in the sense that it can change in response to our experiences, connecting different networks in different ways. This facilitates the combination of information.

Such areas where lots of synergy takes placemostly in the the front and middle of the cortex (the brains outer layer)integrate different sources of information from the entire brain. They are therefore more widely and efficiently connected with the rest of the brain than the regions which deal with primary sensory and movement-related information.

High-synergy areas that support integration of information also typically have lots of synapses, the microscopic connections that enable nerve cells to communicate.

We wanted to know whether this ability to accumulate and build information through complex networks across the brain is different between humans and other primates, which are close relatives of ours in evolutionary terms.

To find out, we looked at brain imaging data and genetic analyses of different species. We found that synergistic interactions account for a higher proportion of total information flow in the human brain than in the brains of macaque monkeys. In contrast, the brains of both species are equal in terms of how much they rely on redundant information.

However, we also looked specifically at the prefrontal cortex, an area in the front of the brain that supports more advanced cognitive functioning. In macaques, redundant information processing is more prevalent in this region, whereas in humans it is a synergy-heavy area.

The prefrontal cortex has also undergone significant expansion with evolution. When we examined data from chimpanzee brains, we found that the more a region of the human brain had expanded during evolution in size relative to its counterpart in the chimp, the more this region relied on synergy.

We also looked at genetic analyses from human donors. This showed that brain regions associated with processing synergistic information are more likely to express genes that are uniquely human and related to brain development and function, such as intelligence.

This led us to the conclusion that additional human brain tissue, acquired as a result of evolution, may be primarily dedicated to synergy. In turn, it is tempting to speculate that the advantages of greater synergy may, in part, explain our species additional cognitive capabilities. Synergy may add an important piece to the puzzle of human brain evolution, which was previously missing.

Ultimately, our work reveals how the human brain navigates the trade-off between reliability and integration of information; we need both. Importantly, the framework we developed holds the promise of critical new insights into a wide array of neuroscientific questions, from those about general cognition to disorders.

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Image Credit: Gerrit Bril from Pixabay

See more here:

What Is It About the Human Brain That Makes Us Smarter Than Other Animals? New Research - Singularity Hub