Tag Archives: nd
#437964 How Explainable Artificial Intelligence ...
The field of artificial intelligence has created computers that can drive cars, synthesize chemical compounds, fold proteins, and detect high-energy particles at a superhuman level.
However, these AI algorithms cannot explain the thought processes behind their decisions. A computer that masters protein folding and also tells researchers more about the rules of biology is much more useful than a computer that folds proteins without explanation.
Therefore, AI researchers like me are now turning our efforts toward developing AI algorithms that can explain themselves in a manner that humans can understand. If we can do this, I believe that AI will be able to uncover and teach people new facts about the world that have not yet been discovered, leading to new innovations.
Learning From Experience
One field of AI, called reinforcement learning, studies how computers can learn from their own experiences. In reinforcement learning, an AI explores the world, receiving positive or negative feedback based on its actions.
This approach has led to algorithms that have independently learned to play chess at a superhuman level and prove mathematical theorems without any human guidance. In my work as an AI researcher, I use reinforcement learning to create AI algorithms that learn how to solve puzzles such as the Rubik’s Cube.
Through reinforcement learning, AIs are independently learning to solve problems that even humans struggle to figure out. This has got me and many other researchers thinking less about what AI can learn and more about what humans can learn from AI. A computer that can solve the Rubik’s Cube should be able to teach people how to solve it, too.
Peering Into the Black Box
Unfortunately, the minds of superhuman AIs are currently out of reach to us humans. AIs make terrible teachers and are what we in the computer science world call “black boxes.”
AI simply spits out solutions without giving reasons for its solutions. Computer scientists have been trying for decades to open this black box, and recent research has shown that many AI algorithms actually do think in ways that are similar to humans. For example, a computer trained to recognize animals will learn about different types of eyes and ears and will put this information together to correctly identify the animal.
The effort to open up the black box is called explainable AI. My research group at the AI Institute at the University of South Carolina is interested in developing explainable AI. To accomplish this, we work heavily with the Rubik’s Cube.
The Rubik’s Cube is basically a pathfinding problem: Find a path from point A—a scrambled Rubik’s Cube—to point B—a solved Rubik’s Cube. Other pathfinding problems include navigation, theorem proving and chemical synthesis.
My lab has set up a website where anyone can see how our AI algorithm solves the Rubik’s Cube; however, a person would be hard-pressed to learn how to solve the cube from this website. This is because the computer cannot tell you the logic behind its solutions.
Solutions to the Rubik’s Cube can be broken down into a few generalized steps—the first step, for example, could be to form a cross while the second step could be to put the corner pieces in place. While the Rubik’s Cube itself has over 10 to the 19th power possible combinations, a generalized step-by-step guide is very easy to remember and is applicable in many different scenarios.
Approaching a problem by breaking it down into steps is often the default manner in which people explain things to one another. The Rubik’s Cube naturally fits into this step-by-step framework, which gives us the opportunity to open the black box of our algorithm more easily. Creating AI algorithms that have this ability could allow people to collaborate with AI and break down a wide variety of complex problems into easy-to-understand steps.
A step-by-step refinement approach can make it easier for humans to understand why AIs do the things they do. Forest Agostinelli, CC BY-ND
Collaboration Leads to Innovation
Our process starts with using one’s own intuition to define a step-by-step plan thought to potentially solve a complex problem. The algorithm then looks at each individual step and gives feedback about which steps are possible, which are impossible and ways the plan could be improved. The human then refines the initial plan using the advice from the AI, and the process repeats until the problem is solved. The hope is that the person and the AI will eventually converge to a kind of mutual understanding.
Currently, our algorithm is able to consider a human plan for solving the Rubik’s Cube, suggest improvements to the plan, recognize plans that do not work and find alternatives that do. In doing so, it gives feedback that leads to a step-by-step plan for solving the Rubik’s Cube that a person can understand. Our team’s next step is to build an intuitive interface that will allow our algorithm to teach people how to solve the Rubik’s Cube. Our hope is to generalize this approach to a wide range of pathfinding problems.
People are intuitive in a way unmatched by any AI, but machines are far better in their computational power and algorithmic rigor. This back and forth between man and machine utilizes the strengths from both. I believe this type of collaboration will shed light on previously unsolved problems in everything from chemistry to mathematics, leading to new solutions, intuitions and innovations that may have, otherwise, been out of reach.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
Image Credit: Serg Antonov / Unsplash Continue reading
Posted in Human Robots
#437171 Scientists Tap the World’s Most ...
In The Hitchhiker’s Guide to the Galaxy by Douglas Adams, the haughty supercomputer Deep Thought is asked whether it can find the answer to the ultimate question concerning life, the universe, and everything. It replies that, yes, it can do it, but it’s tricky and it’ll have to think about it. When asked how long it will take it replies, “Seven-and-a-half million years. I told you I’d have to think about it.”
Real-life supercomputers are being asked somewhat less expansive questions but tricky ones nonetheless: how to tackle the Covid-19 pandemic. They’re being used in many facets of responding to the disease, including to predict the spread of the virus, to optimize contact tracing, to allocate resources and provide decisions for physicians, to design vaccines and rapid testing tools, and to understand sneezes. And the answers are needed in a rather shorter time frame than Deep Thought was proposing.
The largest number of Covid-19 supercomputing projects involves designing drugs. It’s likely to take several effective drugs to treat the disease. Supercomputers allow researchers to take a rational approach and aim to selectively muzzle proteins that SARS-CoV-2, the virus that causes Covid-19, needs for its life cycle.
The viral genome encodes proteins needed by the virus to infect humans and to replicate. Among these are the infamous spike protein that sniffs out and penetrates its human cellular target, but there are also enzymes and molecular machines that the virus forces its human subjects to produce for it. Finding drugs that can bind to these proteins and stop them from working is a logical way to go.
The Summit supercomputer at Oak Ridge National Laboratory has a peak performance of 200,000 trillion calculations per second—equivalent to about a million laptops. Image credit: Oak Ridge National Laboratory, U.S. Dept. of Energy, CC BY
I am a molecular biophysicist. My lab, at the Center for Molecular Biophysics at the University of Tennessee and Oak Ridge National Laboratory, uses a supercomputer to discover drugs. We build three-dimensional virtual models of biological molecules like the proteins used by cells and viruses, and simulate how various chemical compounds interact with those proteins. We test thousands of compounds to find the ones that “dock” with a target protein. Those compounds that fit, lock-and-key style, with the protein are potential therapies.
The top-ranked candidates are then tested experimentally to see if they indeed do bind to their targets and, in the case of Covid-19, stop the virus from infecting human cells. The compounds are first tested in cells, then animals, and finally humans. Computational drug discovery with high-performance computing has been important in finding antiviral drugs in the past, such as the anti-HIV drugs that revolutionized AIDS treatment in the 1990s.
World’s Most Powerful Computer
Since the 1990s the power of supercomputers has increased by a factor of a million or so. Summit at Oak Ridge National Laboratory is presently the world’s most powerful supercomputer, and has the combined power of roughly a million laptops. A laptop today has roughly the same power as a supercomputer had 20-30 years ago.
However, in order to gin up speed, supercomputer architectures have become more complicated. They used to consist of single, very powerful chips on which programs would simply run faster. Now they consist of thousands of processors performing massively parallel processing in which many calculations, such as testing the potential of drugs to dock with a pathogen or cell’s proteins, are performed at the same time. Persuading those processors to work together harmoniously is a pain in the neck but means we can quickly try out a lot of chemicals virtually.
Further, researchers use supercomputers to figure out by simulation the different shapes formed by the target binding sites and then virtually dock compounds to each shape. In my lab, that procedure has produced experimentally validated hits—chemicals that work—for each of 16 protein targets that physician-scientists and biochemists have discovered over the past few years. These targets were selected because finding compounds that dock with them could result in drugs for treating different diseases, including chronic kidney disease, prostate cancer, osteoporosis, diabetes, thrombosis and bacterial infections.
Scientists are using supercomputers to find ways to disable the various proteins—including the infamous spike protein (green protrusions)—produced by SARS-CoV-2, the virus responsible for Covid-19. Image credit: Thomas Splettstoesser scistyle.com, CC BY-ND
Billions of Possibilities
So which chemicals are being tested for Covid-19? A first approach is trying out drugs that already exist for other indications and that we have a pretty good idea are reasonably safe. That’s called “repurposing,” and if it works, regulatory approval will be quick.
But repurposing isn’t necessarily being done in the most rational way. One idea researchers are considering is that drugs that work against protein targets of some other virus, such as the flu, hepatitis or Ebola, will automatically work against Covid-19, even when the SARS-CoV-2 protein targets don’t have the same shape.
Our own work has now expanded to about 10 targets on SARS-CoV-2, and we’re also looking at human protein targets for disrupting the virus’s attack on human cells. Top-ranked compounds from our calculations are being tested experimentally for activity against the live virus. Several of these have already been found to be active.The best approach is to check if repurposed compounds will actually bind to their intended target. To that end, my lab published a preliminary report of a supercomputer-driven docking study of a repurposing compound database in mid-February. The study ranked 8,000 compounds in order of how well they bind to the viral spike protein. This paper triggered the establishment of a high-performance computing consortium against our viral enemy, announced by President Trump in March. Several of our top-ranked compounds are now in clinical trials.
Also, we and others are venturing out into the wild world of new drug discovery for Covid-19—looking for compounds that have never been tried as drugs before. Databases of billions of these compounds exist, all of which could probably be synthesized in principle but most of which have never been made. Billion-compound docking is a tailor-made task for massively parallel supercomputing.
Dawn of the Exascale Era
Work will be helped by the arrival of the next big machine at Oak Ridge, called Frontier, planned for next year. Frontier should be about 10 times more powerful than Summit. Frontier will herald the “exascale” supercomputing era, meaning machines capable of 1,000,000,000,000,000,000 calculations per second.
Although some fear supercomputers will take over the world, for the time being, at least, they are humanity’s servants, which means that they do what we tell them to. Different scientists have different ideas about how to calculate which drugs work best—some prefer artificial intelligence, for example—so there’s quite a lot of arguing going on.
Hopefully, scientists armed with the most powerful computers in the world will, sooner rather than later, find the drugs needed to tackle Covid-19. If they do, then their answers will be of more immediate benefit, if less philosophically tantalizing, than the answer to the ultimate question provided by Deep Thought, which was, maddeningly, simply 42.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
Image credit: NIH/NIAID Continue reading
Posted in Human Robots
#435308 Brain-Machine Interfaces Are Getting ...
Elon Musk grabbed a lot of attention with his July 16 announcement that his company Neuralink plans to implant electrodes into the brains of people with paralysis by next year. Their first goal is to create assistive technology to help people who can’t move or are unable to communicate.
If you haven’t been paying attention, brain-machine interfaces (BMIs) that allow people to control robotic arms with their thoughts might sound like science fiction. But science and engineering efforts have already turned it into reality.
In a few research labs around the world, scientists and physicians have been implanting devices into the brains of people who have lost the ability to control their arms or hands for over a decade. In our own research group at the University of Pittsburgh, we’ve enabled people with paralyzed arms and hands to control robotic arms that allow them to grasp and move objects with relative ease. They can even experience touch-like sensations from their own hand when the robot grasps objects.
At its core, a BMI is pretty straightforward. In your brain, microscopic cells called neurons are sending signals back and forth to each other all the time. Everything you think, do and feel as you interact with the world around you is the result of the activity of these 80 billion or so neurons.
If you implant a tiny wire very close to one of these neurons, you can record the electrical activity it generates and send it to a computer. Record enough of these signals from the right area of the brain and it becomes possible to control computers, robots, or anything else you might want, simply by thinking about moving. But doing this comes with tremendous technical challenges, especially if you want to record from hundreds or thousands of neurons.
What Neuralink Is Bringing to the Table
Elon Musk founded Neuralink in 2017, aiming to address these challenges and raise the bar for implanted neural interfaces.
Perhaps the most impressive aspect of Neuralink’s system is the breadth and depth of their approach. Building a BMI is inherently interdisciplinary, requiring expertise in electrode design and microfabrication, implantable materials, surgical methods, electronics, packaging, neuroscience, algorithms, medicine, regulatory issues, and more. Neuralink has created a team that spans most, if not all, of these areas.
With all of this expertise, Neuralink is undoubtedly moving the field forward, and improving their technology rapidly. Individually, many of the components of their system represent significant progress along predictable paths. For example, their electrodes, that they call threads, are very small and flexible; many researchers have tried to harness those properties to minimize the chance the brain’s immune response would reject the electrodes after insertion. Neuralink has also developed high-performance miniature electronics, another focus area for labs working on BMIs.
Often overlooked in academic settings, however, is how an entire system would be efficiently implanted in a brain.
Neuralink’s BMI requires brain surgery. This is because implanted electrodes that are in intimate contact with neurons will always outperform non-invasive electrodes where neurons are far away from the electrodes sitting outside the skull. So, a critical question becomes how to minimize the surgical challenges around getting the device into a brain.
Maybe the most impressive aspect of Neuralink’s announcement was that they created a 3,000-electrode neural interface where electrodes could be implanted at a rate of between 30 and 200 per minute. Each thread of electrodes is implanted by a sophisticated surgical robot that essentially acts like a sewing machine. This all happens while specifically avoiding blood vessels that blanket the surface of the brain. The robotics and imaging that enable this feat, with tight integration to the entire device, is striking.
Neuralink has thought through the challenge of developing a clinically viable BMI from beginning to end in a way that few groups have done, though they acknowledge that many challenges remain as they work towards getting this technology into human patients in the clinic.
Figuring Out What More Electrodes Gets You
The quest for implantable devices with thousands of electrodes is not only the domain of private companies. DARPA, the NIH BRAIN Initiative, and international consortiums are working on neurotechnologies for recording and stimulating in the brain with goals of tens of thousands of electrodes. But what might scientists do with the information from 1,000, 3,000, or maybe even 100,000 neurons?
At some level, devices with more electrodes might not actually be necessary to have a meaningful impact in people’s lives. Effective control of computers for access and communication, of robotic limbs to grasp and move objects as well as of paralyzed muscles is already happening—in people. And it has been for a number of years.
Since the 1990s, the Utah Array, which has just 100 electrodes and is manufactured by Blackrock Microsystems, has been a critical device in neuroscience and clinical research. This electrode array is FDA-cleared for temporary neural recording. Several research groups, including our own, have implanted Utah Arrays in people that lasted multiple years.
Currently, the biggest constraints are related to connectors, electronics, and system-level engineering, not the implanted electrode itself—although increasing the electrodes’ lifespan to more than five years would represent a significant advance. As those technical capabilities improve, it might turn out that the ability to accurately control computers and robots is limited more by scientists’ understanding of what the neurons are saying—that is, the neural code—than by the number of electrodes on the device.
Even the most capable implanted system, and maybe the most capable devices researchers can reasonably imagine, might fall short of the goal of actually augmenting skilled human performance. Nevertheless, Neuralink’s goal of creating better BMIs has the potential to improve the lives of people who can’t move or are unable to communicate. Right now, Musk’s vision of using BMIs to meld physical brains and intelligence with artificial ones is no more than a dream.
So, what does the future look like for Neuralink and other groups creating implantable BMIs? Devices with more electrodes that last longer and are connected to smaller and more powerful wireless electronics are essential. Better devices themselves, however, are insufficient. Continued public and private investment in companies and academic research labs, as well as innovative ways for these groups to work together to share technologies and data, will be necessary to truly advance scientists’ understanding of the brain and deliver on the promise of BMIs to improve peoples’ lives.
While researchers need to keep the future societal implications of advanced neurotechnologies in mind—there’s an essential role for ethicists and regulation—BMIs could be truly transformative as they help more people overcome limitations caused by injury or disease in the brain and body.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
Image Credit: UPMC/Pitt Health Sciences, / CC BY-NC-ND Continue reading
Posted in Human Robots