Tag Archives: thinking

#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

#435224 Can AI Save the Internet from Fake News?

There’s an old proverb that says “seeing is believing.” But in the age of artificial intelligence, it’s becoming increasingly difficult to take anything at face value—literally.

The rise of so-called “deepfakes,” in which different types of AI-based techniques are used to manipulate video content, has reached the point where Congress held its first hearing last month on the potential abuses of the technology. The congressional investigation coincided with the release of a doctored video of Facebook CEO Mark Zuckerberg delivering what appeared to be a sinister speech.

View this post on Instagram

‘Imagine this…’ (2019) Mark Zuckerberg reveals the truth about Facebook and who really owns the future… see more @sheffdocfest VDR technology by @cannyai #spectreknows #privacy #democracy #surveillancecapitalism #dataism #deepfake #deepfakes #contemporaryartwork #digitalart #generativeart #newmediaart #codeart #markzuckerberg #artivism #contemporaryart

A post shared by Bill Posters (@bill_posters_uk) on Jun 7, 2019 at 7:15am PDT

Scientists are scrambling for solutions on how to combat deepfakes, while at the same time others are continuing to refine the techniques for less nefarious purposes, such as automating video content for the film industry.

At one end of the spectrum, for example, researchers at New York University’s Tandon School of Engineering have proposed implanting a type of digital watermark using a neural network that can spot manipulated photos and videos.

The idea is to embed the system directly into a digital camera. Many smartphone cameras and other digital devices already use AI to boost image quality and make other corrections. The authors of the study out of NYU say their prototype platform increased the chances of detecting manipulation from about 45 percent to more than 90 percent without sacrificing image quality.

On the other hand, researchers at Carnegie Mellon University recently hit on a technique for automatically and rapidly converting large amounts of video content from one source into the style of another. In one example, the scientists transferred the facial expressions of comedian John Oliver onto the bespectacled face of late night show host Stephen Colbert.

The CMU team says the method could be a boon to the movie industry, such as by converting black and white films to color, though it also conceded that the technology could be used to develop deepfakes.

Words Matter with Fake News
While the current spotlight is on how to combat video and image manipulation, a prolonged trench warfare on fake news is being fought by academia, nonprofits, and the tech industry.

This isn’t the fake news that some have come to use as a knee-jerk reaction to fact-based information that might be less than flattering to the subject of the report. Rather, fake news is deliberately-created misinformation that is spread via the internet.

In a recent Pew Research Center poll, Americans said fake news is a bigger problem than violent crime, racism, and terrorism. Fortunately, many of the linguistic tools that have been applied to determine when people are being deliberately deceitful can be baked into algorithms for spotting fake news.

That’s the approach taken by a team at the University of Michigan (U-M) to develop an algorithm that was better than humans at identifying fake news—76 percent versus 70 percent—by focusing on linguistic cues like grammatical structure, word choice, and punctuation.

For example, fake news tends to be filled with hyperbole and exaggeration, using terms like “overwhelming” or “extraordinary.”

“I think that’s a way to make up for the fact that the news is not quite true, so trying to compensate with the language that’s being used,” Rada Mihalcea, a computer science and engineering professor at U-M, told Singularity Hub.

The paper “Automatic Detection of Fake News” was based on the team’s previous studies on how people lie in general, without necessarily having the intention of spreading fake news, she said.

“Deception is a complicated and complex phenomenon that requires brain power,” Mihalcea noted. “That often results in simpler language, where you have shorter sentences or shorter documents.”

AI Versus AI
While most fake news is still churned out by humans with identifiable patterns of lying, according to Mihalcea, other researchers are already anticipating how to detect misinformation manufactured by machines.

A group led by Yejin Choi, with the Allen Institute of Artificial Intelligence and the University of Washington in Seattle, is one such team. The researchers recently introduced the world to Grover, an AI platform that is particularly good at catching autonomously-generated fake news because it’s equally good at creating it.

“This is due to a finding that is perhaps counterintuitive: strong generators for neural fake news are themselves strong detectors of it,” wrote Rowan Zellers, a PhD student and team member, in a Medium blog post. “A generator of fake news will be most familiar with its own peculiarities, such as using overly common or predictable words, as well as the peculiarities of similar generators.”

The team found that the best current discriminators can classify neural fake news from real, human-created text with 73 percent accuracy. Grover clocks in with 92 percent accuracy based on a training set of 5,000 neural network-generated fake news samples. Zellers wrote that Grover got better at scale, identifying 97.5 percent of made-up machine mumbo jumbo when trained on 80,000 articles.

It performed almost as well against fake news created by a powerful new text-generation system called GPT-2 built by OpenAI, a nonprofit research lab founded by Elon Musk, classifying 96.1 percent of the machine-written articles.

OpenAI had so feared that the platform could be abused that it has only released limited versions of the software. The public can play with a scaled-down version posted by a machine learning engineer named Adam King, where the user types in a short prompt and GPT-2 bangs out a short story or poem based on the snippet of text.

No Silver AI Bullet
While real progress is being made against fake news, the challenges of using AI to detect and correct misinformation are abundant, according to Hugo Williams, outreach manager for Logically, a UK-based startup that is developing different detectors using elements of deep learning and natural language processing, among others. He explained that the Logically models analyze information based on a three-pronged approach.

Publisher metadata: Is the article from a known, reliable, and trustworthy publisher with a history of credible journalism?
Network behavior: Is the article proliferating through social platforms and networks in ways typically associated with misinformation?
Content: The AI scans articles for hundreds of known indicators typically found in misinformation.

“There is no single algorithm which is capable of doing this,” Williams wrote in an email to Singularity Hub. “Even when you have a collection of different algorithms which—when combined—can give you relatively decent indications of what is unreliable or outright false, there will always need to be a human layer in the pipeline.”

The company released a consumer app in India back in February just before that country’s election cycle that was a “great testing ground” to refine its technology for the next app release, which is scheduled in the UK later this year. Users can submit articles for further scrutiny by a real person.

“We see our technology not as replacing traditional verification work, but as a method of simplifying and streamlining a very manual process,” Williams said. “In doing so, we’re able to publish more fact checks at a far quicker pace than other organizations.”

“With heightened analysis and the addition of more contextual information around the stories that our users are reading, we are not telling our users what they should or should not believe, but encouraging critical thinking based upon reliable, credible, and verified content,” he added.

AI may never be able to detect fake news entirely on its own, but it can help us be smarter about what we read on the internet.

Image Credit: Dennis Lytyagin / Shutterstock.com Continue reading

Posted in Human Robots

#435098 Coming of Age in the Age of AI: The ...

The first generation to grow up entirely in the 21st century will never remember a time before smartphones or smart assistants. They will likely be the first children to ride in self-driving cars, as well as the first whose healthcare and education could be increasingly turned over to artificially intelligent machines.

Futurists, demographers, and marketers have yet to agree on the specifics of what defines the next wave of humanity to follow Generation Z. That hasn’t stopped some, like Australian futurist Mark McCrindle, from coining the term Generation Alpha, denoting a sort of reboot of society in a fully-realized digital age.

“In the past, the individual had no power, really,” McCrindle told Business Insider. “Now, the individual has great control of their lives through being able to leverage this world. Technology, in a sense, transformed the expectations of our interactions.”

No doubt technology may impart Marvel superhero-like powers to Generation Alpha that even tech-savvy Millennials never envisioned over cups of chai latte. But the powers of machine learning, computer vision, and other disciplines under the broad category of artificial intelligence will shape this yet unformed generation more definitively than any before it.

What will it be like to come of age in the Age of AI?

The AI Doctor Will See You Now
Perhaps no other industry is adopting and using AI as much as healthcare. The term “artificial intelligence” appears in nearly 90,000 publications from biomedical literature and research on the PubMed database.

AI is already transforming healthcare and longevity research. Machines are helping to design drugs faster and detect disease earlier. And AI may soon influence not only how we diagnose and treat illness in children, but perhaps how we choose which children will be born in the first place.

A study published earlier this month in NPJ Digital Medicine by scientists from Weill Cornell Medicine used 12,000 photos of human embryos taken five days after fertilization to train an AI algorithm on how to tell which in vitro fertilized embryo had the best chance of a successful pregnancy based on its quality.

Investigators assigned each embryo a grade based on various aspects of its appearance. A statistical analysis then correlated that grade with the probability of success. The algorithm, dubbed Stork, was able to classify the quality of a new set of images with 97 percent accuracy.

“Our algorithm will help embryologists maximize the chances that their patients will have a single healthy pregnancy,” said Dr. Olivier Elemento, director of the Caryl and Israel Englander Institute for Precision Medicine at Weill Cornell Medicine, in a press release. “The IVF procedure will remain the same, but we’ll be able to improve outcomes by harnessing the power of artificial intelligence.”

Other medical researchers see potential in applying AI to detect possible developmental issues in newborns. Scientists in Europe, working with a Finnish AI startup that creates seizure monitoring technology, have developed a technique for detecting movement patterns that might indicate conditions like cerebral palsy.

Published last month in the journal Acta Pediatrica, the study relied on an algorithm to extract the movements from a newborn, turning it into a simplified “stick figure” that medical experts could use to more easily detect clinically relevant data.

The researchers are continuing to improve the datasets, including using 3D video recordings, and are now developing an AI-based method for determining if a child’s motor maturity aligns with its true age. Meanwhile, a study published in February in Nature Medicine discussed the potential of using AI to diagnose pediatric disease.

AI Gets Classy
After being weaned on algorithms, Generation Alpha will hit the books—about machine learning.

China is famously trying to win the proverbial AI arms race by spending billions on new technologies, with one Chinese city alone pledging nearly $16 billion to build a smart economy based on artificial intelligence.

To reach dominance by its stated goal of 2030, Chinese cities are also incorporating AI education into their school curriculum. Last year, China published its first high school textbook on AI, according to the South China Morning Post. More than 40 schools are participating in a pilot program that involves SenseTime, one of the country’s biggest AI companies.

In the US, where it seems every child has access to their own AI assistant, researchers are just beginning to understand how the ubiquity of intelligent machines will influence the ways children learn and interact with their highly digitized environments.

Sandra Chang-Kredl, associate professor of the department of education at Concordia University, told The Globe and Mail that AI could have detrimental effects on learning creativity or emotional connectedness.

Similar concerns inspired Stefania Druga, a member of the Personal Robots group at the MIT Media Lab (and former Education Teaching Fellow at SU), to study interactions between children and artificial intelligence devices in order to encourage positive interactions.

Toward that goal, Druga created Cognimates, a platform that enables children to program and customize their own smart devices such as Alexa or even a smart, functional robot. The kids can also use Cognimates to train their own AI models or even build a machine learning version of Rock Paper Scissors that gets better over time.

“I believe it’s important to also introduce young people to the concepts of AI and machine learning through hands-on projects so they can make more informed and critical use of these technologies,” Druga wrote in a Medium blog post.

Druga is also the founder of Hackidemia, an international organization that sponsors workshops and labs around the world to introduce kids to emerging technologies at an early age.

“I think we are in an arms race in education with the advancement of technology, and we need to start thinking about AI literacy before patterns of behaviors for children and their families settle in place,” she wrote.

AI Goes Back to School
It also turns out that AI has as much to learn from kids. More and more researchers are interested in understanding how children grasp basic concepts that still elude the most advanced machine minds.

For example, developmental psychologist Alison Gopnik has written and lectured extensively about how studying the minds of children can provide computer scientists clues on how to improve machine learning techniques.

In an interview on Vox, she described that while DeepMind’s AlpahZero was trained to be a chessmaster, it struggles with even the simplest changes in the rules, such as allowing the bishop to move horizontally instead of vertically.

“A human chess player, even a kid, will immediately understand how to transfer that new rule to their playing of the game,” she noted. “Flexibility and generalization are something that even human one-year-olds can do but that the best machine learning systems have a much harder time with.”

Last year, the federal defense agency DARPA announced a new program aimed at improving AI by teaching it “common sense.” One of the chief strategies is to develop systems for “teaching machines through experience, mimicking the way babies grow to understand the world.”

Such an approach is also the basis of a new AI program at MIT called the MIT Quest for Intelligence.

The research leverages cognitive science to understand human intelligence, according to an article on the project in MIT Technology Review, such as exploring how young children visualize the world using their own innate 3D models.

“Children’s play is really serious business,” said Josh Tenenbaum, who leads the Computational Cognitive Science lab at MIT and his head of the new program. “They’re experiments. And that’s what makes humans the smartest learners in the known universe.”

In a world increasingly driven by smart technologies, it’s good to know the next generation will be able to keep up.

Image Credit: phoelixDE / Shutterstock.com Continue reading

Posted in Human Robots

#435056 How Researchers Used AI to Better ...

A few years back, DeepMind’s Demis Hassabis famously prophesized that AI and neuroscience will positively feed into each other in a “virtuous circle.” If realized, this would fundamentally expand our insight into intelligence, both machine and human.

We’ve already seen some proofs of concept, at least in the brain-to-AI direction. For example, memory replay, a biological mechanism that fortifies our memories during sleep, also boosted AI learning when abstractly appropriated into deep learning models. Reinforcement learning, loosely based on our motivation circuits, is now behind some of AI’s most powerful tools.

Hassabis is about to be proven right again.

Last week, two studies independently tapped into the power of ANNs to solve a 70-year-old neuroscience mystery: how does our visual system perceive reality?

The first, published in Cell, used generative networks to evolve DeepDream-like images that hyper-activate complex visual neurons in monkeys. These machine artworks are pure nightmare fuel to the human eye; but together, they revealed a fundamental “visual hieroglyph” that may form a basic rule for how we piece together visual stimuli to process sight into perception.

In the second study, a team used a deep ANN model—one thought to mimic biological vision—to synthesize new patterns tailored to control certain networks of visual neurons in the monkey brain. When directly shown to monkeys, the team found that the machine-generated artworks could reliably activate predicted populations of neurons. Future improved ANN models could allow even better control, giving neuroscientists a powerful noninvasive tool to study the brain. The work was published in Science.

The individual results, though fascinating, aren’t necessarily the point. Rather, they illustrate how scientists are now striving to complete the virtuous circle: tapping AI to probe natural intelligence. Vision is only the beginning—the tools can potentially be expanded into other sensory domains. And the more we understand about natural brains, the better we can engineer artificial ones.

It’s a “great example of leveraging artificial intelligence to study organic intelligence,” commented Dr. Roman Sandler at Kernel.co on Twitter.

Why Vision?
ANNs and biological vision have quite the history.

In the late 1950s, the legendary neuroscientist duo David Hubel and Torsten Wiesel became some of the first to use mathematical equations to understand how neurons in the brain work together.

In a series of experiments—many using cats—the team carefully dissected the structure and function of the visual cortex. Using myriads of images, they revealed that vision is processed in a hierarchy: neurons in “earlier” brain regions, those closer to the eyes, tend to activate when they “see” simple patterns such as lines. As we move deeper into the brain, from the early V1 to a nub located slightly behind our ears, the IT cortex, neurons increasingly respond to more complex or abstract patterns, including faces, animals, and objects. The discovery led some scientists to call certain IT neurons “Jennifer Aniston cells,” which fire in response to pictures of the actress regardless of lighting, angle, or haircut. That is, IT neurons somehow extract visual information into the “gist” of things.

That’s not trivial. The complex neural connections that lead to increasing abstraction of what we see into what we think we see—what we perceive—is a central question in machine vision: how can we teach machines to transform numbers encoding stimuli into dots, lines, and angles that eventually form “perceptions” and “gists”? The answer could transform self-driving cars, facial recognition, and other computer vision applications as they learn to better generalize.

Hubel and Wiesel’s Nobel-prize-winning studies heavily influenced the birth of ANNs and deep learning. Much of earlier ANN “feed-forward” model structures are based on our visual system; even today, the idea of increasing layers of abstraction—for perception or reasoning—guide computer scientists to build AI that can better generalize. The early romance between vision and deep learning is perhaps the bond that kicked off our current AI revolution.

It only seems fair that AI would feed back into vision neuroscience.

Hieroglyphs and Controllers
In the Cell study, a team led by Dr. Margaret Livingstone at Harvard Medical School tapped into generative networks to unravel IT neurons’ complex visual alphabet.

Scientists have long known that neurons in earlier visual regions (V1) tend to fire in response to “grating patches” oriented in certain ways. Using a limited set of these patches like letters, V1 neurons can “express a visual sentence” and represent any image, said Dr. Arash Afraz at the National Institute of Health, who was not involved in the study.

But how IT neurons operate remained a mystery. Here, the team used a combination of genetic algorithms and deep generative networks to “evolve” computer art for every studied neuron. In seven monkeys, the team implanted electrodes into various parts of the visual IT region so that they could monitor the activity of a single neuron.

The team showed each monkey an initial set of 40 images. They then picked the top 10 images that stimulated the highest neural activity, and married them to 30 new images to “evolve” the next generation of images. After 250 generations, the technique, XDREAM, generated a slew of images that mashed up contorted face-like shapes with lines, gratings, and abstract shapes.

This image shows the evolution of an optimum image for stimulating a visual neuron in a monkey. Image Credit: Ponce, Xiao, and Schade et al. – Cell.
“The evolved images look quite counter-intuitive,” explained Afraz. Some clearly show detailed structures that resemble natural images, while others show complex structures that can’t be characterized by our puny human brains.

This figure shows natural images (right) and images evolved by neurons in the inferotemporal cortex of a monkey (left). Image Credit: Ponce, Xiao, and Schade et al. – Cell.
“What started to emerge during each experiment were pictures that were reminiscent of shapes in the world but were not actual objects in the world,” said study author Carlos Ponce. “We were seeing something that was more like the language cells use with each other.”

This image was evolved by a neuron in the inferotemporal cortex of a monkey using AI. Image Credit: Ponce, Xiao, and Schade et al. – Cell.
Although IT neurons don’t seem to use a simple letter alphabet, it does rely on a vast array of characters like hieroglyphs or Chinese characters, “each loaded with more information,” said Afraz.

The adaptive nature of XDREAM turns it into a powerful tool to probe the inner workings of our brains—particularly for revealing discrepancies between biology and models.

The Science study, led by Dr. James DiCarlo at MIT, takes a similar approach. Using ANNs to generate new patterns and images, the team was able to selectively predict and independently control neuron populations in a high-level visual region called V4.

“So far, what has been done with these models is predicting what the neural responses would be to other stimuli that they have not seen before,” said study author Dr. Pouya Bashivan. “The main difference here is that we are going one step further and using the models to drive the neurons into desired states.”

It suggests that our current ANN models for visual computation “implicitly capture a great deal of visual knowledge” which we can’t really describe, but which the brain uses to turn vision information into perception, the authors said. By testing AI-generated images on biological vision, however, the team concluded that today’s ANNs have a degree of understanding and generalization. The results could potentially help engineer even more accurate ANN models of biological vision, which in turn could feed back into machine vision.

“One thing is clear already: Improved ANN models … have led to control of a high-level neural population that was previously out of reach,” the authors said. “The results presented here have likely only scratched the surface of what is possible with such implemented characterizations of the brain’s neural networks.”

To Afraz, the power of AI here is to find cracks in human perception—both our computational models of sensory processes, as well as our evolved biological software itself. AI can be used “as a perfect adversarial tool to discover design cracks” of IT, said Afraz, such as finding computer art that “fools” a neuron into thinking the object is something else.

“As artificial intelligence researchers develop models that work as well as the brain does—or even better—we will still need to understand which networks are more likely to behave safely and further human goals,” said Ponce. “More efficient AI can be grounded by knowledge of how the brain works.”

Image Credit: Sangoiri / Shutterstock.com Continue reading

Posted in Human Robots

#434837 In Defense of Black Box AI

Deep learning is powering some amazing new capabilities, but we find it hard to scrutinize the workings of these algorithms. Lack of interpretability in AI is a common concern and many are trying to fix it, but is it really always necessary to know what’s going on inside these “black boxes”?

In a recent perspective piece for Science, Elizabeth Holm, a professor of materials science and engineering at Carnegie Mellon University, argued in defense of the black box algorithm. I caught up with her last week to find out more.

Edd Gent: What’s your experience with black box algorithms?

Elizabeth Holm: I got a dual PhD in materials science and engineering and scientific computing. I came to academia about six years ago and part of what I wanted to do in making this career change was to refresh and revitalize my computer science side.

I realized that computer science had changed completely. It used to be about algorithms and making codes run fast, but now it’s about data and artificial intelligence. There are the interpretable methods like random forest algorithms, where we can tell how the machine is making its decisions. And then there are the black box methods, like convolutional neural networks.

Once in a while we can find some information about their inner workings, but most of the time we have to accept their answers and kind of probe around the edges to figure out the space in which we can use them and how reliable and accurate they are.

EG: What made you feel like you had to mount a defense of these black box algorithms?

EH: When I started talking with my colleagues, I found that the black box nature of many of these algorithms was a real problem for them. I could understand that because we’re scientists, we always want to know why and how.

It got me thinking as a bit of a contrarian, “Are black boxes all bad? Must we reject them?” Surely not, because human thought processes are fairly black box. We often rely on human thought processes that the thinker can’t necessarily explain.

It’s looking like we’re going to be stuck with these methods for a while, because they’re really helpful. They do amazing things. And so there’s a very pragmatic realization that these are the best methods we’ve got to do some really important problems, and we’re not right now seeing alternatives that are interpretable. We’re going to have to use them, so we better figure out how.

EG: In what situations do you think we should be using black box algorithms?

EH: I came up with three rules. The simplest rule is: when the cost of a bad decision is small and the value of a good decision is high, it’s worth it. The example I gave in the paper is targeted advertising. If you send an ad no one wants it doesn’t cost a lot. If you’re the receiver it doesn’t cost a lot to get rid of it.

There are cases where the cost is high, and that’s then we choose the black box if it’s the best option to do the job. Things get a little trickier here because we have to ask “what are the costs of bad decisions, and do we really have them fully characterized?” We also have to be very careful knowing that our systems may have biases, they may have limitations in where you can apply them, they may be breakable.

But at the same time, there are certainly domains where we’re going to test these systems so extensively that we know their performance in virtually every situation. And if their performance is better than the other methods, we need to do it. Self driving vehicles are a significant example—it’s almost certain they’re going to have to use black box methods, and that they’re going to end up being better drivers than humans.

The third rule is the more fun one for me as a scientist, and that’s the case where the black box really enlightens us as to a new way to look at something. We have trained a black box to recognize the fracture energy of breaking a piece of metal from a picture of the broken surface. It did a really good job, and humans can’t do this and we don’t know why.

What the computer seems to be seeing is noise. There’s a signal in that noise, and finding it is very difficult, but if we do we may find something significant to the fracture process, and that would be an awesome scientific discovery.

EG: Do you think there’s been too much emphasis on interpretability?

EH: I think the interpretability problem is a fundamental, fascinating computer science grand challenge and there are significant issues where we need to have an interpretable model. But how I would frame it is not that there’s too much emphasis on interpretability, but rather that there’s too much dismissiveness of uninterpretable models.

I think that some of the current social and political issues surrounding some very bad black box outcomes have convinced people that all machine learning and AI should be interpretable because that will somehow solve those problems.

Asking humans to explain their rationale has not eliminated bias, or stereotyping, or bad decision-making in humans. Relying too much on interpreted ability perhaps puts the responsibility in the wrong place for getting better results. I can make a better black box without knowing exactly in what way the first one was bad.

EG: Looking further into the future, do you think there will be situations where humans will have to rely on black box algorithms to solve problems we can’t get our heads around?

EH: I do think so, and it’s not as much of a stretch as we think it is. For example, humans don’t design the circuit map of computer chips anymore. We haven’t for years. It’s not a black box algorithm that designs those circuit boards, but we’ve long since given up trying to understand a particular computer chip’s design.

With the billions of circuits in every computer chip, the human mind can’t encompass it, either in scope or just the pure time that it would take to trace every circuit. There are going to be cases where we want a system so complex that only the patience that computers have and their ability to work in very high-dimensional spaces is going to be able to do it.

So we can continue to argue about interpretability, but we need to acknowledge that we’re going to need to use black boxes. And this is our opportunity to do our due diligence to understand how to use them responsibly, ethically, and with benefits rather than harm. And that’s going to be a social conversation as well as as a scientific one.

*Responses have been edited for length and style

Image Credit: Chingraph / Shutterstock.com Continue reading

Posted in Human Robots