Category Archives: Singularity

Deep learning has transformed the field of artificial intelligence, but the limitations of conventional computer hardware are already hindering progress. Researchers at MIT think their new nanophotonic processor could be the answer by carrying out deep learning at the speed of light.

In the 1980s, scientists and engineers hailed optical computing as the next great revolution in information technology, but it turned out that bulky components like fiber optic cables and lenses didnt make for particularly robust or compact computers.

In particular, they found it extremely challenging to make scalable optical logic gates, and therefore impractical to make general optical computers, according to MIT physics post-doc Yichen Shen. One thing light is good at, though, is multiplying matricesarrays of numbers arranged in columns and rows. You can actually mathematically explain the way a lens acts on a beam of light in terms of matrix multiplications.

This also happens to be a core component of the calculations involved in deep learning. Combined with advances in nanophotonicsthe study of lights behavior at the nanometer scalethis has led to a resurgence in interest in optical computing.

Deep learning is mainly matrix multiplications, so it works very well with the nature of light, says Shen. With light you can make deep learning computing much faster and thousands of times more energy-efficient.

To demonstrate this, Shen and his MIT colleagues have designed an all-optical chip that can implement artificial neural networksthe brain-inspired algorithms at the heart of deep learning.

In a recent paper in Nature, the group describes a chip made up of 56 interferometerscomponents that allow the researchers to control how beams of light interfere with each other to carry out mathematical operations.

The processor can be reprogrammed by applying a small voltage to the waveguides that direct beams of light around the processor, which heats them and causes them to change shape.

The chip is best suited to inference tasks, the researchers say, where the algorithm is put to practical use by applying a learned model to analyze new data, for instance to detect objects in an image.

It isnt great at learning, because heating the waveguides is relatively slow compared to how electronic systems are reprogrammed. So, in their study, the researchers trained the algorithm on a computer before transferring the learned model to the nanophotonic processor to carry out the inference task.

Thats not a major issue. For many practical applications its not necessary to carry out learning and inference on the same chip. Google recently made headlines for designing its own deep learning chip, the TPU, which is also specifically designed for inference and most companies that use a lot of machine learning split the two jobs.

In many cases they update these models once every couple of months and the rest of the time the fixed model is just doing inference, says Shen. People usually separate these tasks. They typically have a server just doing training and another just doing inference, so I dont see a big problem making a chip focused on inference.

Once the model has been programmed into the chip, it can then carry out computations at the speed of light using less than one-thousandth the energy per operation compared to conventional electronic chips.

There are limitations, though. Because the chip deals with light waves that operate on the scale of a few microns, there are fundamental limits to how small these chips can get.

"The wavelength really sets the limit of how small the waveguides can be. We wont be able to make devices significantly smaller. Maybe by a factor of four, but physics will ultimately stop us, says MIT graduate student Nicholas Harris, who co-authored the paper.

That means it would be difficult to implement neural nets much larger than a few thousand neurons. However, the vast majority of current deep learning algorithms are well within that limit.

The system did achieve a significantly lower accuracy on the task than a standard computer implementing the same deep learning model, correctly identifying 76.7 percent of vowels compared to 91.7 percent.

But Harris says they think this was largely due to interference between the various heating elements used to program the waveguides, and that it should be easy to fix by using thermal isolation trenches or extra calibration steps.

Importantly, the chips are also built using the same fabrication technology as conventional computer chips, so scaling up production should be easy. Shen said the group has already had interest in their technology from prominent chipmakers.

Pierre-Alexandre Blanche, a professor of optics at the University of Arizona, said hes very excited by the paper, which he said complements his own work. But he cautioned against getting too carried away.

This is another milestone in the progress toward useful optical computing. But we are still far away to be competitive with electronics, he told Singularity Hub in an email. The argumentation about scalability, power consumption, speed etc. [in the paper] use a lot of conditional tense and assumptions which demonstrate that, if there is potential indeed, there is still a lot of research to be done.

In particular, he pointed out that the system was only a partial solution to the problem. While the vast majority of neuronal computation involves multiplication of matrices, there is another component: calculating a non-linear response.

In the current paper this aspect of the computation was simulated on a regular computer. The researchers say in future models this function could be carried out by a so-called saturable absorber integrated into the waveguides that absorbs less light as the intensity increases.

But Blanche notes that this is not a trivial problem and something his group is actually currently working on. It is not like you can buy one at the drug store, he says. Bhavin Shastri, a post-doc at Princeton whose group is also working on nanophotonic chips for implementing neural networks, said the research was important, as enabling matrix multiplications is a key step to enabling full-fledged photonic neural networks.

Overall, this area of research is poised to usher in an exciting and promising field, he added. Neural networks implemented in photonic hardware could revolutionize how machines interact with ultrafast physical phenomena. Silicon photonics combines the analog device performance of photonics with the cost and scalability of silicon manufacturing.

Stock media provided by across/Pond5.com

See the rest here:

Deep Learning at the Speed of Light on Nanophotonic Chips - Singularity Hub

Windows Central

Ashes of the Singularity: Escalation v2.3 is a great reason to return to the battlefield Windows Central Dreadnoughts are powerful assets in Ashes of the Singularity, but now there's something more devastating to be unleashed on the battlefield. Update 2.3 for the Escalation expansion adds a new class of ship, called "Juggernauts." These are colossal ...

Read the rest here:

Ashes of the Singularity: Escalation v2.3 is a great reason to return to the battlefield - Windows Central

As Bill Gates sees it, there are three main threats to our species: nuclear war, climate change, and the next global pandemic.

Speaking on pandemic preparedness at the Munich Security Conference earlier this year, Gates reminded us that the fact that a deadly global pandemic has not occurred in recent history shouldnt be mistaken for evidence that a deadly pandemic will not occur in the future.

If we want to be prepared for the worst, Gates says, first and most importantly, we have to build an arsenal of new weaponsvaccines, drugs, and diagnostics.

Some scientists are now using computers to do just that.

Despite the availability of the flu shot, the World Health Organization reports that seasonal influenza is still responsible for millions of serious illnesses and as many as half a million deaths per year globally. The partial efficacy of each years flu shot, coupled with long manufacturing times and limited global availability, suggests new flu-fighting methods are still needed.

And thats just for the seasonal flu. Pandemic influenza, like the devastating 1918 Spanish flu, could again kill tens of millions of people in a single year.

Antibodies, a natural part of the immune system, are front-line soldiers in the war against viruses. The job of an antibody is to recognize and physically adhere to a foreign invader like influenza. Human antibodies are bivalent, meaning they have two hands with which they can grab onto their target.

Under a microscope, influenza looks like a tiny ball with spikes. It uses some of its surface spikes to break into human cells. By grabbing tightly to those spikes using one or both hands, antibodies can prevent flu particles from infecting human cells. But every year the rapidly evolving influenza picks up mutations in its spike proteins, causing the sticky hands of our antibodies to no longer recognize the virus.

Researchers have long sought a universal flu vaccineone that doesnt need to be readministered every year. Efforts to produce one tend to involve injecting noninfectious flu lookalikes in hopes that it will prime the immune system to mount a proper attack on whatever real strain of flu it sees next. Despite some progress, researchers have not yet been able to coax the immune system to defend against all strains of influenza, and the threat of a global pandemic still looms.

Transmission electron microscopic image of an influenza virus particle. Image credit: CDC/ Erskine. L. Palmer, Ph.D.; M. L. Martin

Computational protein design offers another way. Rather than relying on the immune system to generate an antibody protein capable of shutting down a virus like the flu, computer modeling can now help quickly create custom antiviral proteins programmed to shut down a deadly virus.

Unlike a vaccine, this class of drug could be administered to treat an existing infection or given days prior to exposure to prevent one. And because these designer proteins work independently of the immune system, their potency does not depend on having an intact immune systema useful trait, as those with weaker immune systems are at high risk for viral infection.

Computer-generated antiviral proteins work the same way some natural proteins in our immune system do. By having surfaces that are chemically complementary to their targets, antiviral proteins can stick tightly to a specific virus. If a protein sticks to a virus in just the right way, it can physically block how that virus moves, ultimately preventing infection.

By designing an antiviral protein on a computer, building it in the laboratory, and then administering it into the body, you effectively digitize part of the immune system.

In 2016, computer-generated proteins were shown to be more effective than oseltamivir (Tamiflu) in warding off death in influenza-infected mice. One dose of designer protein given intranasally was more effective than 10 doses of Tamiflu, a drug considered an essential medicine by the WHO due to its antiflu activity. Whats more, these new computer-generated antiflu proteins protected mice against diverse strains of the flu. Efforts to turn these promising results into FDA-approved drugs are underway.

In a just-published paper in Nature Biotechnology, scientists here at the Institute for Protein Design at the University of Washington went a step further and demonstrated a new way to shut down the flu: They used computer modeling to build a completely new kind of antiviral protein with three sticky hands.

Why three? It turns out many deadly envelope viruseslike influenza, Ebola, and HIVbuild their spike proteins out of three symmetric parts.

A single antiviral drug with three properly spaced hands should be able to symmetrically grab each part of a spike protein, leading to tighter binding and overall better antiviral activity. This geometric feat is beyond what the human immune system can naturally do.

Left: The tips of many viral spike proteins are built out of three symmetric parts, with one part highlighted in pink. Right: A new three-handed antiflu protein (blue) bound to influenzas HA spike.Image Credit: UW Institute for Protein Design, CC BY-ND

The design strategy worked. The best three-handed protein, called Tri-HSB.1C, was able to bind tightly to diverse strains of influenza. When given to mice, it also afforded complete protection against a lethal flu infection with only minimal associated weight lossa trait commonly used to diagnose flu severity in mice. Researchers are now applying the same tools to the Ebola spike protein.

It will be many years before this new technology is approved for use in humans for any virus. But we may not have to wait long to see some lifesaving benefits.

By coating a strip of paper with a three-handed flu binder and applying influenza samples on top, the same team was able to detect the presence of viral surface protein even at very low concentrations. This proof-of-concept detection system could be transformed into a reliable and affordable on-site diagnostic tool for a variety of viruses by detecting them in saliva or blood. Like a pregnancy test, a band on a test strip could indicate flu. Or Ebola. Or the next rapidly spreading global pandemic.

In a 2015 letter to the New England Journal of Medicine on lessons learned from the Ebola epidemic in West Africa, Bill Gates describes the lack of preparation by the global community as a global failure.

Perhaps the only good news from the tragic Ebola epidemic, Gates says, is that it may serve as a wake-up call. (The Bill and Melinda Gates Foundation funds work on protein design at the University of Washington.)

When a global viral pandemic like the 1918 Spanish flu strikes again, antivirus software of the biological kind may play an important role in saving millions of lives.

This article was originally published on The Conversation. Read the original article.

Disclosure statement: Ian Haydon is a doctoral student at the University of Washington's Institute for Protein Design, which receives funding from the Bill and Melinda Gates foundation.

Read the rest here:

Designing Antiviral Proteins via Computer Could Help Halt the Next Pandemic - Singularity Hub

Announcement: A new major release, RDK 2.0, is now available! Download source code or a bootable ISO at the Releases tab, or retrieve the latest Source Code from the repository at the Source Code tab.

Project Description

The Singularity Research Development Kit (RDK) is based on the Microsoft Research Singularity Project. It includes source code, build tools, test suites, design notes, and other background materials. The Singularity RDK is for academic non-commercial use and is governed by this license.

About Singularity

Singularity is a research project focused on the construction of dependable systems through innovation in the areas of systems, languages, and tools. We are building a research operating system prototype (called Singularity), extending programming languages, and developing new techniques and tools for specifying and verifying program behavior.

Advances in languages, compilers, and tools open the possibility of significantly improving software. For example, Singularity uses type-safe languages and an abstract instruction set to enable what we call Software Isolated Processes (SIPs). SIPs provide the strong isolation guarantees of OS processes (isolated object space, separate GCs, separate runtimes) without the overhead of hardware-enforced protection domains. In the current Singularity prototype SIPs are extremely cheap; they run in ring 0 in the kernels address space.

Singularity uses these advances to build more reliable systems and applications. For example, because SIPs are so cheap to create and enforce, Singularity runs each program, device driver, or system extension in its own SIP. SIPs are not allowed to share memory or modify their own code. As a result, we can make strong reliability guarantees about the code running in a SIP. We can verify much broader properties about a SIP at compile or install time than can be done for code running in traditional OS processes. Broader application of static verification is critical to predicting system behavior and providing users with strong guarantees about reliability.

See also: Singularity: Rethinking Dependable System Design Singularity: Rethinking the Software Stack Using the Singularity Research Development Kit

Read the original:

Singularity RDK - Home

Scientists, technologists, engineers, and visionaries are building the future. Amazing things are in the pipeline. Its a big deal. But you already knew all that. Such speculation is common. Whats less common? Scale.

How big is big?

Silicon Valley, Silicon Alley, Silicon Dock, all of the Silicons around the world, they are dreaming the dream. They are innovating, Catherine Wood said at Singularity Universitys Exponential Finance in New York. We are sizing the opportunity. That's what we do.

Wood is founder and CEO of ARK Investment Management, a research and investment company focused on the growth potential of todays disruptive technologies. Prior to ARK, she served as CIO of Global ThematicStrategies at AllianceBernstein for 12 years.

We believe innovation is key to growth, Wood said. We are not focused on the past. We are focused on the future. We think there are tremendous opportunities in the public marketplace because this shift towards passive [investing] has created a lot of risk aversion and tremendous inefficiencies.

In a new research report, released this week, ARK took a look at seven disruptive technologies, and put a number on just how tremendous they are. Heres what they found.

(Check out ARKs website and free report, Big Ideas of 2017, for more numbers, charts, and detail.)

Deep learning is a subcategory of machine learning which is itself a subcategory of artificial intelligence. Deep learning is the source of much of the hype surrounding AI today. (You know you may be in a hype bubble when ads tout AI on Sunday golf commercial breaks.)

Behind the hype, however, big tech companies are pursuing deep learningto do very practical things. And whereas the internet, which unleashed trillions in market value, transformedseveralindustriesnews, entertainment, advertising, etc.deep learning will work its way intoeven more, Wood said.

As deep learning advances, it shouldautomate and improve technology, transportation, manufacturing, healthcare, finance, and more. And as is often the case with emerging technologies, it may form entirely new businesses we have yet to imagine.

Bill Gates has said a breakthrough in machine learning would be worth 10 Microsofts. Microsoft is $550 to $600 billion, Wood said. We think deep learning is going to be twice that. We think [it] could approach $17 trillion in market capwhich would be 35 Amazons.

Wood didnt mince words about a future when cars drive themselves.

This is the biggest change that the automotive industry has ever faced, she said.

Todays automakers have a global market capitalization of a trillion dollars. Meanwhile, mobility-as-a-service companies as a whole (think ridesharing) are valued around $115 billion. If this number took into account expectations of a driverless future, itd be higher.

The mobility-as-a-service market, which will slash the cost of "point-to-point" travel, couldbe worth more than todays automakers combined, Wood said. Twice as much, in fact. As gross sales grow to something like $10 trillion in the early 2030s, her firm thinks some 20% of that will go to platform providers. It could be a $2 trillion opportunity.

Wood said a handful of companies will dominate the market, and Tesla is well positioned to be one of those companies. They are developing both the hardware, electric cars, and the software, self-driving algorithms. And although analysts tend to look at them as a just an automaker right now, thats not all theyll be down the road.

We think if [Tesla] got even 5% of this global market for autonomous taxi networks, it should be worth another $100 billion today, Wood said.

3D printing has become part of mainstream consciousness thanks, mostly, to the prospect of desktop printers for consumer prices. But these are imperfect, and the dream of an at-home replicator still eludes us. The manufacturing industry, however, is much closer to using 3D printers at scale.

Not long ago, we wrote about Carbons partnership with Adidas to mass-produce shoe midsoles. This is significant because, whereas industrial 3D printing has focused on prototyping to date, improvingcost, quality, and speed are makingitviable for finished products.

According to ARK, 3D printing may grow into a $41 billion market by 2020, and Wood noteda McKinsey forecast of as much as $490 billion by 2025. McKinsey will be right if 3D printing actually becomes a part of the industrial production process, so end-use parts, Wood said.

According to ARK, the cost of genome editing has fallen 28x to 52x (depending on reagents) in the last four years. CRISPR is the technique leading the genome editing revolution, dramatically cutting time and cost while maintaining editing efficiency. Despite its potential, Wood said she isnt hearing enough about it from investors yet.

There are roughly 10,000 monogenic or single-gene diseases. Only 5% are treatable today, she said. ARK believes treating these diseases is worth an annual $70 billion globally. Other areas of interest include stem cell therapy research, personalized medicine, drug development, agriculture, biofuels, and more.

Still,the big names in this areaIntellia, Editas, and CRISPRarent on the radar.

You can see if a company in this space has a strong IP position, as Genentech did in 1980, then the growth rates can be enormous, Wood said. Again, you don't hear these names, and that's quite interesting to me. We think there are very low expectations in that space.

By 2020, 75% of the world will own a smartphone, according to ARK. Amid smartphones' many uses, mobile payments will be one of the most impactful. Coupled with better security (biometrics) and wider acceptance (NFC and point-of-sale), ARK thinks mobile transactions couldgrow 15x, from $1 trillion today to upwards of $15 trillion by2020.

In addition, to making sharing economy transactions more frictionless, they are generally keyto financial inclusion in emergingand developed markets, ARK says. And big emerging markets, such as India and China, are at the forefront, thanks to favorable regulations.

Asia is leading the charge here, Wood said. You look at companies like Tencent and Alipay. They are really moving very quickly towards mobile and actually showing us the way.

Robots arent just for auto manufacturers anymore. Driven by continued cost declines and easier programming, more businessesare adopting robots.Amazons robot workforce in warehouses has grown from 1,000 to nearly 50,000 since 2014. And they have never laid off anyone, other than for performance reasons, in their distribution centers, Wood said.

But she understands fears over lost jobs.

This is only the beginning of a big round of automation driven by cheaper, smarter, safer, and more flexible robots. She agrees there will be a lot of displacement. Still, some commentatorsoverlook associated productivity gains. By 2035, Wood said US GDP couldbe $12 trillion more than it would have been without robotics and automationthats a $40 trillion economy instead of a $28 trillion economy.

This is the history of technology. Productivity. New products and services. It is our job as investors to figure out where that $12 trillion is, Wood said. We can't even imagine it right now. We couldn't imagine what the internet was going to do with us in the early '90s.

Blockchain-enabled cryptoassets, such as Bitcoin, Ethereum, and Steem, have caused more than a stir in recent years. In addition to Bitcoin, there are now some 700 cryptoassets of various shapes and hues. Bitcoin still rules the roostwitha market value of nearly $40 billion, up from just $3 billion two years ago, according to ARK. But its only half the total.

This market is nascent. There are a lot of growing pains taking place right now in the crypto world, but the promise is there, Wood said. Its a very hot space.

Like all young markets, ARK says, cryptoasset markets are characterized by enthusiasm, uncertainty, and speculation. The firms blockchain products lead, Chris Burniske, uses Twitterwhich is where he says the communitycongregatesto take the temperature. In a recent Twitter poll, 62% of respondents said they believed the markets total value would exceed a trillion dollars in 10 years. In a followup, more focused on thetrillion-plus crowd, 35% favored$1$5 trillion, 17% guessed $5$10 trillion, and 34% chose $10+ trillion.

Looking pastthe speculation, Wood believes theres at least one bigarea blockchain and cryptoassets are poised to break into: the $500-billion, fee-based business of sending money across borders known as remittances.

If you look at the Philippines-to-South Korean corridor, what you're seeing already is that Bitcoin is 20% of the remittances market, Wood said. The migrant workers who are transmitting currency, they don't know that Bitcoin is what's enabling such a low-fee transaction. It's the rails, effectively. They just see the fiat transfer. We think that that's going to be a very exciting market.

Stock media provided by NomadSoul1/Pond5.com

See more here:

These 7 Disruptive Technologies Could Be Worth Trillions of Dollars - Singularity Hub

Singularity University (SU), a global community using exponential technologies to tackle the worlds greatest challenges, will hold its first international summit on the African continent. The two-day SingularityU South Africa Summit is being hosted in collaboration with Standard Bank, and with key strategic partners, including Deloitte, MTN, 702, and SAP and is being produced by Mann Made Media. SingularityU South Africa Summit will convene exponential thought leaders, SU faculty, and organizations from around the world to provide participants with insights into emerging exponential technologies and how they can be used to create positive change and economic growth in the region. Corporate South Africa realises the importance of change and the influence of innovation and technology across all sectors. In response, this unique summit in Johannesburg will present a display of advanced technologies, extensive debate, and collaborative discussions, offering an exchange of ideas and existing best practices in the fields of healthcare, cyberspace, AI, robotics, big data, finance, and design. In addition to expert presentations, participants will explore questions ranging from trending technological changes across the globe, to their impact on industry growth and region-specific challenges. The Summit will also showcase African entrepreneurs and innovations in the interactive exhibitor halls. Singularity University is proud to be working with Standard Bank and Mann Made Media to host this first-ever SingularityU South Africa Summit, and to connect with Africas leaders and organizations shaping the future, says Rob Nail, associate founder and CEO of Singularity University. South Africa represents a microcosm of the challenges facing humanity worldwide and is fast gaining a solid reputation as a global centre. Through this Summit, we hope to connect and inspire leaders in the region to effect global impact. SingularityU Summits are two-day conferences held around the globe to help local leaders understand how exponential technologies can be used to create positive change and economic growth in their region. Summits become an annual point of contact and inspiration for the local community, a catalyst to accelerate a local culture of innovation, and an opportunity to highlight breakthrough technologies, startups, and ideas. SingularityU Summits are attended by the general public, government officials, entrepreneurs, investors, NGOs, impact partners, and educators, and may include educational tracks for government and youth.

Read more here:

Singularity Summit comes to SA - IT-Online

Picture this: youre sitting in a police interrogation room, struggling to describe the face of a criminal to a sketch artist. You pause, wrinkling your brow, trying to remember the distance between his eyes and the shape of his nose.

Suddenly, the detective offers you an easier way: would you like to have your brain scanned instead, so that machines can automatically reconstruct the face in your mind's eye from reading your brain waves?

Sound fantastical? Its not. After decades of work, scientists at Caltech may have finally cracked our brains facial recognition code. Using brain scans and direct neuron recording from macaque monkeys, the team found specialized face patches that respond to specific combinations of facial features.

Like dials on a music mixer, each patch is fine-tuned to a particular set of visual information, which then channel together in different combinations to form a holistic representation of every distinctive face.

The values of each dial were so predictable that scientists were able to recreate a face the monkey saw simply by recording the electrical activity of roughly 200 brain cells. When placed together, the reconstruction and the actual photo were nearly indistinguishable.

This was mind-blowing, says lead author Dr. Doris Tsao.

Even more incredibly, the work completely kills the dominant theory of facial processing, potentially ushering in a revolution in neuroscience, says Dr. Rodrigo Quian Quiroga, a neuroscientist at the University of Leichester who was not involved in the work.

On average, humans are powerful face detectors, beating even the most sophisticated face-tagging algorithms.

Most of us are equipped with the uncanny ability to spot a familiar set of features from a crowd of eyes, noses and mouths. We can unconsciously process a new face in milliseconds, andwhen exposed to that face over and overoften retain that memory for decades to come.

Under the hood, however, facial recognition is anything but simple. Why is it that we can detect a face under dim lighting, half obscured or at a weird angle, but machines cant? What makes peoples faces distinctively their own?

When light reflected off a face hits your retina, the information passes through several layers of neurons before it reaches a highly specialized region of the visual cortex: the inferotemporal (IT) region, a small nugget of brain at the base of the brain. This region is home to face cells: groups of neurons that only respond to faces but not to objects such as houses or landscapes.

In the early 2000s, while recording from epilepsy patients with electrodes implanted into their brains, Quian Quiroga and colleagues found that face cells are particularly picky. So-called Jennifer Aniston cells, for example, would only fire in response to photos of her face and her face alone. The cells quietly ignored all other images, including those of her with Brad Pitt.

This led to a prevailing theory that still dominates the field: that the brain contains specialized face neurons that only respond to one or a few faces, but do so holistically.

But theres a problem: the theory doesnt explain how we process new faces, nor does it get into the nitty-gritty of how faces are actually encoded inside those neurons.

In a stroke of luck, Tsao and team blew open the black box of facial recognition while working on a different problem: how to describe a face mathematically, with a matrix of numbers.

Using a set of 200 faces from an online database, the team first identified landmark features and labeled them with dots. This created a large set of abstract dot-to-dot faces, similar to what filmmakers do during motion capture.

Then, using a statistical method called principle component analysis, the scientists extracted 25 measurements that best represented a given face. These measurements were mostly holistic: one shape dimension, for example, encodes for the changes in hairline, face width, and height of eyes.

By varying these shape dimensions, the authors generated a set of 2,000 black-and-white faces with slight differences in the distance between the brows, skin texture, and other facial features.

In macaque monkeys with electrodes implanted into their brains, the team recorded from three face patchesbrain areas that respond especially to faceswhile showing the monkeys the computer-generated faces.

As it turns out, each face neuron only cared about a single set of features. A neuron that only cares about hairline and skinny eyebrows, for example, would fire up when it detects variations in those features across faces. If two faces had similar hairlines but different mouths, those hairline neurons stayed silent.

Whats more, cells in different face patches processed complementary information. The anterior medial face patch, for example, mainly responded to distances between features (what the team dubs appearance). Other patches fired up to information about shapes, such as the curvature of the nose or length of the mouth.

In a way, these feature neurons are like compasses: they only activate when the measurement is off from a set point (magnetic north, for a compass). Scientists arent quite sure how each cell determines its set point. However, combining all the set points generates a face spacea sort of average face, or a face atlas.

From there, when presented with a new face, each neuron will measure the difference between a feature (distance between eyes, for example) and the face atlas. Combine all those differences, and voilyou have a representation of a new face.

Once the team figured out this division of labor, they constructed a mathematical model to predict how the patches process new faces.

Heres the cool part: the medley of features that best covered the entire shape and look of a face was fairly abstract, including the distance between the brows. Sound familiar? Thats because the brains preferred set of features were similar to the landmarks that the team first intuitively labeled to generate their face database.

We thought we had picked it out of the blue, says Tsao.

But it makes sense. If you look at methods for modeling faces in computer vision, almost all of them...separate out the shape and appearance, she explains. The mathematical elegance of the system is amazing.

The team showed the monkeys a series of new faces while recording from roughly 200 neurons in the face patches. Using their mathematical model, they then calculated what features each neuron encodes for and how they combine.

The result? A stunning accurate reconstruction of the faces the monkeys were seeing. So accurate, in fact, that the algorithm-generated faces were nearly indistinguishable from the original.

It really speaks to how compact and efficient this feature-based neural code is,says Tsao, referring to the fact that such a small set of neurons contained sufficient information for a full face.

Tsaos work doesnt paint the full picture. The team only recorded from two out of six face patches, suggesting that other types of information processing may be happening alongside Tsaos model.

But the study breaks the black box norm thats plagued the field for decades.

Our results demonstrate that at least one subdivision of IT cortex can be explained by an explicit, simple model, and black box explanations are unnecessary, the authors conclude (pretty sassy for an academic paper!).

While there arent any immediate applications, the new findings could eventually guide the development of brain-machine interfaces that directly stimulate the visual cortex to give back sight to the blind. They could also help physicians understand why some people suffer from face blindness, and engineer prosthetics to help.

Image Credit: Doris Tsao

See original here:

Forget Police Sketches: Researchers Perfectly Reconstruct Faces by Reading Brainwaves - Singularity Hub

Its 1998 in Silicon Valley, and PayPal is born.

Many argue this was the moment that launched fintech as we know it. Today, fintech is comprised of roughly 15,000 startups globally, all focused on either enabling or disrupting the industry.

Fintech is still relatively new, and yet it has a remarkable amount of money flowing through it. A recent report from CB Insights found that VC-backed fintech companies raised $2.7 billion in the first quarter of 2017 alone. And the report says the global value of fintechs 22 unicorn companies amounts to $77 billion. While the pace of investment is likely to drop in the US this year, Europe saw an early spike in Q1.

The landscape is rich in opportunity for both investors and startups, from new lending, crowdfunding, and financial management platforms to novel payment, insurance, and investing services.

This week at Singularity Universitys Exponential Finance Summit in New York, Mike Sigal, partner at 500 Startups, gave the audience a snapshot of the current abundance in fintech and a look into how investors and entrepreneurs are viewing the market.

Since it was founded in 2012, 500 Startups remains one of the most active early-stage investors in the world, according to Sigal. The companys made nearly 2,000 seed investments across 50 countries and has $330 million in capital invested.

Within fintech specifically, the firm has invested in over 200 companies across 27 countries and invests in almost 40 new companies each year.

Unlike some traditional VC firms that tend to keep a tight focus on a specific industry vertical, 500 Startups prides themselves on maintaining an extremely diversified investment portfolio. So far, the results have been in their favor.

The company has four unicorns (startups valued a billion dollars) in their portfolio, including Credit Karma, Grab, Stripe, and Twilio. They also have invested in 40 companies that are now each valued between $100 million and a billion dollars.

One of the most interesting things about the financial services industry, Sigal said, is that large portions of it remain untouched by digital technology.

Less than one percent of loans, for example, originate online. This means theres a lot of demand for new digital products to transform existing financial services.

Think about just how much more could be done here, Sigal said.

In many ways, transforming financial services is the name of the game.

Over the last few years, theres been as massive shift in which particular companies are controlling customer expectations while using a financial service. In fact, the companies now controlling user expectations are no longer the banks. Instead, Amazon, Apple, Google, Uber, and Facebook have been setting the tone for customer expectations ever since they moved into financial services.

The funny twist is that thirty percent of fintech investments are still coming from banks and insurers. In short, the big guys who are being disrupted are also willing to invest a lot of money in new solutions that could help them stay competitive.

Another huge opportunity in fintech is the three billion new smartphones users projected to enter the market by 2020. Sigal points out that many current financial services cannot serve this new population with their existing offerings.

Sigal recommends a few specific tactics for early-stage investors to use while selecting companies to invest in.

Most importantly, he notes that VCs often pursue white space where theres open market opportunity. Additionally, Sigal advises investors to pattern match by finding companies that are doing a mix of the following:

To wrap things up, Sigal played a game of fintech hot or not to test how well the audience could identify which technologies are hottest to investors today.

Sigal said many of the technologies receiving the most seed investments are the ones with the most practical market applications. For example, technology for sourcing customer data and technology improving how banks sell to new customers.

Though artificial intelligence and blockchain are the craze in Silicon Valley, Sigal explains they arent necessarily the most appealing technologies to investors yet, unless they have a very clear practical market application.

Its hard to say whether the rapid pace of capital flowing into fintech will continue, but for now, it seems extremely promising to both investors and entrepreneurs.

Cheesy or not, in the case of fintech, get it while its hot.

Image Credit: Pond5

View post:

Get It While It's Hot: Why Fintech Is a Goldmine for Investors - Singularity Hub

The Exponential Finance Summit at the New York Marriott Marquis in Times Square, hosted by the Singularity University and started on June 7, is ending today. You might be still in time to catch the live stream. Otherwise, stay tuned for our coverage of the most interesting points.

Exponential Finance will help attendees navigate the rapid pace of change taking place in the financial sector due to exponential technologies, said Will Weisman, Executive Director for Summits at SU. Well share tools that will help participants stay at the forefront and understand where to invest, how to protect their assets, and what it will take to remain competitive and successful in this new economy.

Co-founded in 2008 by Ray Kurzweil, now Googles director of engineering, and Peter Diamandis, founder and CEO of the XPrize, the Singularity University (SU) is a unique educational and business community focused on exponential technologies able to tackle the worlds biggest challenges.

Exponentially accelerating technologies like Artificial Intelligence (AI), advanced biotech, quantum computing, and robotics, described for example in the works of SU co-founder Ray Kurzweil, promise to bring faster and faster change to all industries. The SU runs educational programs, conferences, incubators, and labs, to help accelerating corporate and social innovation.

Of course, blockchain technology is a perfect example of exponential acceleration in financial technology (fintech). In 2015 and 2016, blockchain fintech was prominently featured at Exponential Finance. This year as well, the conference program is packed with blockchain-related topics, and starts with a pre-summit technology bootcamp on blockchain and distributed ledger technologies.

Other interesting blockchain-related talks include The Boom Of ICOs & Token Offerings Has Blockchain Tech Unlocked A Pandoras Box For Venture Capitals Disruption, by DLT Education president Robert Schwentker, Blockchain, by BitNation partner Toni Lane Casserly, Blockchain in Insurance, by Deloitte principal Eric Piscini and The Institutes CEO Pete Miller, and Taking Blockchain To Production An Ongoing Perspective, by Nuco CEO Matt Spoke.

Blockchain fintech is accelerating exponentially indeed. But it can be argued that financial applications, directly related to digital money and transactions, represent only the tip of the blockchain iceberg, and the most interesting future applications could be those that leverage the power of distributed ledgers in other industries such as AI, robotics, the Internet of Things (IoT), and self-driving cars.

To have a contract with someone, you need to trust them, and you need to have an enforcement mechanism, noted Amin Toufani, Vice President of Strategic Relations and Director of strategy at SU. Blockchain technology is showing us a path to where you can bypass both. Enforcement is automatic and your can trust you counterparty. What if your car offered some bitcoins to the car in front of you? And if and when the cars clear the lane the contract is settled? Enforcement is automatic, you do not need to trust the cars in front of you.

To me, the potential to play a critical role in visionary developments with a potential to make the world a better place is the main appeal of blockchain technology.

Toufanis opening talk, titled Exonomics, explored unifying themes among disruptive trends in business strategy, financial markets, cryptocurrencies, economic policy, and risk management. Stay tuned for our continued coverage next week.

Picture from Wikimedia Commons.

See the rest here:

At Exponential Finance, the Singularity University Explores Visionary Applications of Blockchains - Crypto Insider (press release) (blog)

Subscribe to PCGamesN on YouTube

Along with new Juggernaut-class ships and various balance adjustments, Ashes of the Singularitys latest update adds a new campaign for dedicated space generals. Named 'Genesis', this new single-player content is fully voiced and free for all owners of Ashes of the Singularity.

Armchair generals should check out our list of the PCs best RTS games.

Genesis features six new missions and continues the Ashes story, as the Substrate and the PHC battle for galactic control. It is the most challenging campaign to date, according to Stardock, with it featuring the new Juggernaut ships. Both factions receive their own Juggernaut, with them providing the huge firepower needed to turn a match around. As they are the most powerful unit in the game, Ashes players will not see Juggernauts in quick matches, due to the large cost of building these massive battle-cruisers. Juggernauts also level indefinitely, with them gaining a 5% bonus to damage and health for every level past level five.

The PHCs Juggernaut is the Leonidas, which comes equipped with a massive beam cannon that can tear through enemy Dreadnoughts. The Substrates Juggernaut is focused on overwhelming the opponent by spawning a constant stream of frigates to harass the enemy, and repair its hull.

Stardock have also gone back and rebalanced much of the old Ashes campaigns to make them more fun to play. Maps have been adjusted to make more sense and expert commanders should find the AI is a lot more challenging on the higher difficulty levels.

Stardock have also added five new maps to Ashes skirmish and multiplayer mode, alongside various quality-of-life improvements to make interstellar warfare that bit more enjoyable. For the full list of bug fixes, balance changes and gameplay additions, check out the official Ashes of the Singularity website.

Ashes of the Singularity: Escalation is available on Steam for $39.99/29.99.

See the original post:

Ashes of the Singularity gets a new fully-voiced campaign - PCGamesN