Tag Archives: day

#437809 Q&A: The Masterminds Behind ...

Illustration: iStockphoto

Getting a car to drive itself is undoubtedly the most ambitious commercial application of artificial intelligence (AI). The research project was kicked into life by the 2004 DARPA Urban Challenge and then taken up as a business proposition, first by Alphabet, and later by the big automakers.

The industry-wide effort vacuumed up many of the world’s best roboticists and set rival companies on a multibillion-dollar acquisitions spree. It also launched a cycle of hype that paraded ever more ambitious deadlines—the most famous of which, made by Alphabet’s Sergei Brin in 2012, was that full self-driving technology would be ready by 2017. Those deadlines have all been missed.

Much of the exhilaration was inspired by the seeming miracles that a new kind of AI—deep learning—was achieving in playing games, recognizing faces, and transliterating voices. Deep learning excels at tasks involving pattern recognition—a particular challenge for older, rule-based AI techniques. However, it now seems that deep learning will not soon master the other intellectual challenges of driving, such as anticipating what human beings might do.

Among the roboticists who have been involved from the start are Gill Pratt, the chief executive officer of Toyota Research Institute (TRI) , formerly a program manager at the Defense Advanced Research Projects Agency (DARPA); and Wolfram Burgard, vice president of automated driving technology for TRI and president of the IEEE Robotics and Automation Society. The duo spoke with IEEE Spectrum’s Philip Ross at TRI’s offices in Palo Alto, Calif.

This interview has been condensed and edited for clarity.

IEEE Spectrum: How does AI handle the various parts of the self-driving problem?

Photo: Toyota

Gill Pratt

Gill Pratt: There are three different systems that you need in a self-driving car: It starts with perception, then goes to prediction, and then goes to planning.

The one that by far is the most problematic is prediction. It’s not prediction of other automated cars, because if all cars were automated, this problem would be much more simple. How do you predict what a human being is going to do? That’s difficult for deep learning to learn right now.

Spectrum: Can you offset the weakness in prediction with stupendous perception?

Photo: Toyota Research Institute for Burgard

Wolfram Burgard

Wolfram Burgard: Yes, that is what car companies basically do. A camera provides semantics, lidar provides distance, radar provides velocities. But all this comes with problems, because sometimes you look at the world from different positions—that’s called parallax. Sometimes you don’t know which range estimate that pixel belongs to. That might make the decision complicated as to whether that is a person painted onto the side of a truck or whether this is an actual person.

With deep learning there is this promise that if you throw enough data at these networks, it’s going to work—finally. But it turns out that the amount of data that you need for self-driving cars is far larger than we expected.

Spectrum: When do deep learning’s limitations become apparent?

Pratt: The way to think about deep learning is that it’s really high-performance pattern matching. You have input and output as training pairs; you say this image should lead to that result; and you just do that again and again, for hundreds of thousands, millions of times.

Here’s the logical fallacy that I think most people have fallen prey to with deep learning. A lot of what we do with our brains can be thought of as pattern matching: “Oh, I see this stop sign, so I should stop.” But it doesn’t mean all of intelligence can be done through pattern matching.

“I asked myself, if all of those cars had automated drive, how good would they have to be to tolerate the number of crashes that would still occur?”
—Gill Pratt, Toyota Research Institute

For instance, when I’m driving and I see a mother holding the hand of a child on a corner and trying to cross the street, I am pretty sure she’s not going to cross at a red light and jaywalk. I know from my experience being a human being that mothers and children don’t act that way. On the other hand, say there are two teenagers—with blue hair, skateboards, and a disaffected look. Are they going to jaywalk? I look at that, you look at that, and instantly the probability in your mind that they’ll jaywalk is much higher than for the mother holding the hand of the child. It’s not that you’ve seen 100,000 cases of young kids—it’s that you understand what it is to be either a teenager or a mother holding a child’s hand.

You can try to fake that kind of intelligence. If you specifically train a neural network on data like that, you could pattern-match that. But you’d have to know to do it.

Spectrum: So you’re saying that when you substitute pattern recognition for reasoning, the marginal return on the investment falls off pretty fast?

Pratt: That’s absolutely right. Unfortunately, we don’t have the ability to make an AI that thinks yet, so we don’t know what to do. We keep trying to use the deep-learning hammer to hammer more nails—we say, well, let’s just pour more data in, and more data.

Spectrum: Couldn’t you train the deep-learning system to recognize teenagers and to assign the category a high propensity for jaywalking?

Burgard: People have been doing that. But it turns out that these heuristics you come up with are extremely hard to tweak. Also, sometimes the heuristics are contradictory, which makes it extremely hard to design these expert systems based on rules. This is where the strength of the deep-learning methods lies, because somehow they encode a way to see a pattern where, for example, here’s a feature and over there is another feature; it’s about the sheer number of parameters you have available.

Our separation of the components of a self-driving AI eases the development and even the learning of the AI systems. Some companies even think about using deep learning to do the job fully, from end to end, not having any structure at all—basically, directly mapping perceptions to actions.

Pratt: There are companies that have tried it; Nvidia certainly tried it. In general, it’s been found not to work very well. So people divide the problem into blocks, where we understand what each block does, and we try to make each block work well. Some of the blocks end up more like the expert system we talked about, where we actually code things, and other blocks end up more like machine learning.

Spectrum: So, what’s next—what new technique is in the offing?

Pratt: If I knew the answer, we’d do it. [Laughter]

Spectrum: You said that if all cars on the road were automated, the problem would be easy. Why not “geofence” the heck out of the self-driving problem, and have areas where only self-driving cars are allowed?

Pratt: That means putting in constraints on the operational design domain. This includes the geography—where the car should be automated; it includes the weather, it includes the level of traffic, it includes speed. If the car is going slow enough to avoid colliding without risking a rear-end collision, that makes the problem much easier. Street trolleys operate with traffic still in some parts of the world, and that seems to work out just fine. People learn that this vehicle may stop at unexpected times. My suspicion is, that is where we’ll see Level 4 autonomy in cities. It’s going to be in the lower speeds.

“We are now in the age of deep learning, and we don’t know what will come after.”
—Wolfram Burgard, Toyota Research Institute

That’s a sweet spot in the operational design domain, without a doubt. There’s another one at high speed on a highway, because access to highways is so limited. But unfortunately there is still the occasional debris that suddenly crosses the road, and the weather gets bad. The classic example is when somebody irresponsibly ties a mattress to the top of a car and it falls off; what are you going to do? And the answer is that terrible things happen—even for humans.

Spectrum: Learning by doing worked for the first cars, the first planes, the first steam boilers, and even the first nuclear reactors. We ran risks then; why not now?

Pratt: It has to do with the times. During the era where cars took off, all kinds of accidents happened, women died in childbirth, all sorts of diseases ran rampant; the expected characteristic of life was that bad things happened. Expectations have changed. Now the chance of dying in some freak accident is quite low because of all the learning that’s gone on, the OSHA [Occupational Safety and Health Administration] rules, UL code for electrical appliances, all the building standards, medicine.

Furthermore—and we think this is very important—we believe that empathy for a human being at the wheel is a significant factor in public acceptance when there is a crash. We don’t know this for sure—it’s a speculation on our part. I’ve driven, I’ve had close calls; that could have been me that made that mistake and had that wreck. I think people are more tolerant when somebody else makes mistakes, and there’s an awful crash. In the case of an automated car, we worry that that empathy won’t be there.

Photo: Toyota

Toyota is using this
Platform 4 automated driving test vehicle, based on the Lexus LS, to develop Level-4 self-driving capabilities for its “Chauffeur” project.

Spectrum: Toyota is building a system called Guardian to back up the driver, and a more futuristic system called Chauffeur, to replace the driver. How can Chauffeur ever succeed? It has to be better than a human plus Guardian!

Pratt: In the discussions we’ve had with others in this field, we’ve talked about that a lot. What is the standard? Is it a person in a basic car? Or is it a person with a car that has active safety systems in it? And what will people think is good enough?

These systems will never be perfect—there will always be some accidents, and no matter how hard we try there will still be occasions where there will be some fatalities. At what threshold are people willing to say that’s okay?

Spectrum: You were among the first top researchers to warn against hyping self-driving technology. What did you see that so many other players did not?

Pratt: First, in my own case, during my time at DARPA I worked on robotics, not cars. So I was somewhat of an outsider. I was looking at it from a fresh perspective, and that helps a lot.

Second, [when I joined Toyota in 2015] I was joining a company that is very careful—even though we have made some giant leaps—with the Prius hybrid drive system as an example. Even so, in general, the philosophy at Toyota is kaizen—making the cars incrementally better every single day. That care meant that I was tasked with thinking very deeply about this thing before making prognostications.

And the final part: It was a new job for me. The first night after I signed the contract I felt this incredible responsibility. I couldn’t sleep that whole night, so I started to multiply out the numbers, all using a factor of 10. How many cars do we have on the road? Cars on average last 10 years, though ours last 20, but let’s call it 10. They travel on an order of 10,000 miles per year. Multiply all that out and you get 10 to the 10th miles per year for our fleet on Planet Earth, a really big number. I asked myself, if all of those cars had automated drive, how good would they have to be to tolerate the number of crashes that would still occur? And the answer was so incredibly good that I knew it would take a long time. That was five years ago.

Burgard: We are now in the age of deep learning, and we don’t know what will come after. We are still making progress with existing techniques, and they look very promising. But the gradient is not as steep as it was a few years ago.

Pratt: There isn’t anything that’s telling us that it can’t be done; I should be very clear on that. Just because we don’t know how to do it doesn’t mean it can’t be done. Continue reading

Posted in Human Robots

#437796 AI Seeks ET: Machine Learning Powers ...

Can artificial intelligence help the search for life elsewhere in the solar system? NASA thinks the answer may be “yes”—and not just on Mars either.

A pilot AI system is now being tested for use on the ExoMars mission that is currently slated to launch in the summer or fall of 2022. The machine-learning algorithms being developed will help science teams decide how to test Martian soil samples to return only the most meaningful data.

For ExoMars, the AI system will only be used back on earth to analyze data gather by the ExoMars rover. But if the system proves to be as useful to the rovers as now suspected, a NASA mission to Saturn’s moon Titan (now scheduled for 2026 launch) could automate the scientific sleuthing process in the field. This mission will rely on the Dragonfly octocopter drone to fly from surface location to surface location through Titan’s dense atmosphere and drill for signs of life there.

The hunt for microbial life in another world’s soil, either as fossilized remnants or as present-day samples, is very challenging, says Eric Lyness, software lead of the NASA Goddard Planetary Environments Lab in Greenbelt, Md. There is of course no precedent to draw upon, because no one has yet succeeded in astrobiology’s holy grail quest.

But that doesn’t mean AI can’t provide substantial assistance. Lyness explained that for the past few years he’d been puzzling over how to automate portions of an exploratory mission’s geochemical investigation, wherever in the solar system the scientific craft may be.

Last year he decided to try machine learning. “So we got some interns,” he said. “People right out of college or in college, who have been studying machine learning. … And they did some amazing stuff. It turned into much more than we expected.” Lyness and his collaborators presented their scientific analysis algorithm at a geochemistry conference last month.

Illustration: ESA

The ExoMars rover, named Rosalind Franklin, will be the first that can drill down to 2-meter depths, where living soil bacteria could possibly be found.

ExoMars’s rover—named Rosalind Franklin, after one of the co-discoverers of DNA—will be the first that can drill down to 2-meter depths, beyond where solar UV light might penetrate and kill any life forms. In other words, ExoMars will be the first Martian craft with the ability to reach soil depths where living soil bacteria could possibly be found.

“We could potentially find forms of life, microbes or other things like that,” Lyness said. However, he quickly added, very little conclusive evidence today exists to suggest that there’s present-day (microbial) life on Mars. (NASA’s Curiosity rover has sent back some inexplicable observations of both methane and molecular oxygen in the Martian atmosphere that could conceivably be a sign of microbial life forms, though non-biological processes could explain these anomalies too.)

Less controversially, the Rosalind Franklin rover’s drill could also turn up fossilized evidence of life in the Martian soil from earlier epochs when Mars was more hospitable.

NASA’s contribution to the joint Russian/European Space Agency ExoMars project is an instrument called a mass spectrometer that will be used to analyze soil samples from the drill cores. Here, Lyness said, is where AI could really provide a helping hand.

Because the Dragonfly drone and possibly a future mission to Jupiter’s moon Europa would be operating in hostile environments with less opportunity for data transmission to Earth, automating a craft’s astrobiological exploration would be practically a requirement

The spectrometer, which studies the mass distribution of ions in a sample of material, works by blasting the drilled soil sample with a laser and then mapping out the atomic masses of the various molecules and portions of molecules that the laser has liberated. The problem is any given mass spectrum could originate from any number of source compounds, minerals and components. Which always makes analyzing a mass spectrum a gigantic puzzle.

Lyness said his group is studying the mineral montmorillonite, a commonplace component of the Martian soil, to see the many ways it might reveal itself in a mass spectrum. Then his team sneaks in an organic compound with the montmorillonite sample to see how that changes the mass spectrometer output.

“It could take a long time to really break down a spectrum and understand why you’re seeing peaks at certain [masses] in the spectrum,” he said. “So anything you can do to point scientists into a direction that says, ‘Don’t worry, I know it’s not this kind of thing or that kind of thing,’ they can more quickly identify what’s in there.”

Lyness said the ExoMars mission will provide a fertile training ground for his team’s as-yet-unnamed AI algorithm. (He said he’s open to suggestions—though, please, no spoof Boaty McBoatface submissions need apply.)

Because the Dragonfly drone and possibly a future astrobiology mission to Jupiter’s moon Europa would be operating in much more hostile environments with much less opportunity for data transmission back and forth to Earth, automating a craft’s astrobiological exploration would be practically a requirement.

All of which points to a future in mid-2030s in which a nuclear-powered octocopter on a moon of Saturn flies from location to location to drill for evidence of life on this tantalizingly bio-possible world. And machine learning will help power the science.

“We should be researching how to make the science instruments smarter,” Lyness said. “If you can make it smarter at the source, especially for planetary exploration, it has huge payoffs.” Continue reading

Posted in Human Robots

#437776 Video Friday: This Terrifying Robot Will ...

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!):

CLAWAR 2020 – August 24-26, 2020 – [Virtual Conference]
ICUAS 2020 – September 1-4, 2020 – Athens, Greece
ICRES 2020 – September 28-29, 2020 – Taipei, Taiwan
IROS 2020 – October 25-29, 2020 – Las Vegas, Nevada
ICSR 2020 – November 14-16, 2020 – Golden, Colorado
Let us know if you have suggestions for next week, and enjoy today's videos.

The Aigency, which created the FitBot launch video below, is “the world’s first talent management resource for robotic personalities.”

Robots will be playing a bigger role in our lives in the future. By learning to speak their language and work with them now, we can make this future better for everybody. If you’re a creator that’s producing content to entertain and educate people, robots can be a part of that. And we can help you. Robotic actors can show up alongside the rest of your actors.

The folks at Aigency have put together a compilation reel of clips they’ve put on TikTok, which is nice of them, because some of us don’t know how to TikTok because we’re old and boring.

Do googly eyes violate the terms and conditions?

[ Aigency ]

Shane Wighton of the “Stuff Made Here” YouTube channel, who you might remember from that robotic basketball hoop, has a new invention: A haircut robot. This is not the the first barber bot, but previous designs typically used hair clippers. Shane wanted his robot to use scissors. Hilarious and terrifying at once.

[ Stuff Made Here ]

Starting in October of 2016, Prof. Charlie Kemp and Henry M. Clever invented a new kind of robot. They named the prototype NewRo. In March of 2017, Prof. Kemp filmed this video of Henry operating NewRo to perform a number of assistive tasks. While visiting the Bay Area for a AAAI Symposium workshop at Stanford, Prof. Kemp showed this video to a select group of people to get advice, including Dr. Aaron Edsinger. In August of 2017, Dr. Edsinger and Dr. Kemp founded Hello Robot Inc. to commercialize this patent pending assistive technology. Hello Robot Inc. licensed the intellectual property (IP) from Georgia Tech. After three years of stealthy effort, Hello Robot Inc. revealed Stretch, a new kind of robot!

[ Georgia Tech ]

NASA’s Ingenuity Mars Helicopter will make history's first attempt at powered flight on another planet next spring. It is riding with the agency's next mission to Mars (the Mars 2020 Perseverance rover) as it launches from Cape Canaveral Air Force Station later this summer. Perseverance, with Ingenuity attached to its belly, will land on Mars February 18, 2021.

[ JPL ]

For humans, it can be challenging to manipulate thin flexible objects like ropes, wires, or cables. But if these problems are hard for humans, they are nearly impossible for robots. As a cable slides between the fingers, its shape is constantly changing, and the robot’s fingers must be constantly sensing and adjusting the cable’s position and motion. A group of researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and from the MIT Department of Mechanical Engineering pursued the task from a different angle, in a manner that more closely mimics us humans. The team’s new system uses a pair of soft robotic grippers with high-resolution tactile sensors (and no added mechanical constraints) to successfully manipulate freely moving cables.

The team observed that it was difficult to pull the cable back when it reached the edge of the finger, because of the convex surface of the GelSight sensor. Therefore, they hope to improve the finger-sensor shape to enhance the overall performance. In the future, they plan to study more complex cable manipulation tasks such as cable routing and cable inserting through obstacles, and they want to eventually explore autonomous cable manipulation tasks in the auto industry.

[ MIT ]

Gripping robots typically have troubles grabbing transparent or shiny objects. A new technique by Carnegie Mellon University relies on color camera system and machine learning to recognize shapes based on color.

[ CMU ]

A new robotic prosthetic leg prototype offers a more natural, comfortable gait while also being quieter and more energy efficient than other designs. The key is the use of new small and powerful motors with fewer gears, borrowed from the space industry. This streamlined technology enables a free-swinging knee and regenerative braking, which charges the battery during use with energy that would typically be dissipated when the foot hits the ground. This feature enables the leg to more than double a typical prosthetic user's walking needs with one charge per day.

[ University of Michigan ]

Thanks Kate!

This year’s Wonder League teams have been put to the test not only with the challenges set forth by Wonder Workshop and Cartoon Network as they look to help the creek kids from Craig of the Creek solve the greatest mystery of all – the quest for the Lost Realm but due to forces outside their control. With a global pandemic displacing many teams from one another due to lockdowns and quarantines, these teams continued to push themselves to find new ways to work together, solve problems, communicate more effectively, and push themselves to complete a journey that they started and refused to give up on. We at Wonder Workshop are humbled and in awe of all these teams have accomplished.

[ Wonder Workshop ]

Thanks Nicole!

Meet Colin Creager, a mechanical engineer at NASA's Glenn Research Center. Colin is focusing on developing tires that can be used on other worlds. These tires use coil springs made of a special shape memory alloy that will let rovers move across sharp jagged rocks or through soft sand on the Moon or Mars.

[ NASA ]

To be presented at IROS this year, “the first on robot collision detection system using low cost microphones.”

[ Rutgers ]

Robot and mechanism designs inspired by the art of Origami have the potential to generate compact, deployable, lightweight morphing structures, as seen in nature, for potential applications in search-and-rescue, aerospace systems, and medical devices. However, it is challenging to obtain actuation that is easily patternable, reversible, and made with a scalable manufacturing process for origami-inspired self-folding machines. In this work, we describe an approach to design reversible self-folding machines using liquid crystal elastomer (LCE), that contracts when heated, as an artificial muscle.

[ UCSD ]

Just in case you need some extra home entertainment, and you’d like cleaner floors at the same time.

[ iRobot ]

Sure, toss it from a drone. Or from orbit. Whatever, it’s squishy!

[ Squishy Robotics ]

The [virtual] RSS conference this week featured an excellent lineup of speakers and panels, and the best part about it being virtual is that you can watch them all at your leisure! Here’s what’s been posted so far:

[ RSS 2020 ]

Lockheed Martin Robotics Seminar: Toward autonomous flying insect-sized robots: recent results in fabrication, design, power systems, control, and sensing with Sawyer Fuller.

[ UMD ]

In this episode of the AI Podcast, Lex interviews Sergey Levine.

[ AI Podcast ] Continue reading

Posted in Human Robots

#437769 Q&A: Facebook’s CTO Is at War With ...

Photo: Patricia de Melo Moreira/AFP/Getty Images

Facebook chief technology officer Mike Schroepfer leads the company’s AI and integrity efforts.

Facebook’s challenge is huge. Billions of pieces of content—short and long posts, images, and combinations of the two—are uploaded to the site daily from around the world. And any tiny piece of that—any phrase, image, or video—could contain so-called bad content.

In its early days, Facebook relied on simple computer filters to identify potentially problematic posts by their words, such as those containing profanity. These automatically filtered posts, as well as posts flagged by users as offensive, went to humans for adjudication.

In 2015, Facebook started using artificial intelligence to cull images that contained nudity, illegal goods, and other prohibited content; those images identified as possibly problematic were sent to humans for further review.

By 2016, more offensive photos were reported by Facebook’s AI systems than by Facebook users (and that is still the case).

In 2018, Facebook CEO Mark Zuckerberg made a bold proclamation: He predicted that within five or ten years, Facebook’s AI would not only look for profanity, nudity, and other obvious violations of Facebook’s policies. The tools would also be able to spot bullying, hate speech, and other misuse of the platform, and put an immediate end to them.

Today, automated systems using algorithms developed with AI scan every piece of content between the time when a user completes a post and when it is visible to others on the site—just fractions of a second. In most cases, a violation of Facebook’s standards is clear, and the AI system automatically blocks the post. In other cases, the post goes to human reviewers for a final decision, a workforce that includes 15,000 content reviewers and another 20,000 employees focused on safety and security, operating out of more than 20 facilities around the world.

In the first quarter of this year, Facebook removed or took other action (like appending a warning label) on more than 9.6 million posts involving hate speech, 8.6 million involving child nudity or exploitation, almost 8 million posts involving the sale of drugs, 2.3 million posts involving bullying and harassment, and tens of millions of posts violating other Facebook rules.

Right now, Facebook has more than 1,000 engineers working on further developing and implementing what the company calls “integrity” tools. Using these systems to screen every post that goes up on Facebook, and doing so in milliseconds, is sucking up computing resources. Facebook chief technology officer Mike Schroepfer, who is heading up Facebook’s AI and integrity efforts, spoke with IEEE Spectrum about the team’s progress on building an AI system that detects bad content.

Since that discussion, Facebook’s policies around hate speech have come under increasing scrutiny, with particular attention on divisive posts by political figures. A group of major advertisers in June announced that they would stop advertising on the platform while reviewing the situation, and civil rights groups are putting pressure on others to follow suit until Facebook makes policy changes related to hate speech and groups that promote hate, misinformation, and conspiracies.

Facebook CEO Mark Zuckerberg responded with news that Facebook will widen the category of what it considers hateful content in ads. Now the company prohibits claims that people from a specific race, ethnicity, national origin, religious affiliation, caste, sexual orientation, gender identity, or immigration status are a threat to the physical safety, health, or survival of others. The policy change also aims to better protect immigrants, migrants, refugees, and asylum seekers from ads suggesting these groups are inferior or expressing contempt. Finally, Zuckerberg announced that the company will label some problematic posts by politicians and government officials as content that violates Facebook’s policies.

However, civil rights groups say that’s not enough. And an independent audit released in July also said that Facebook needs to go much further in addressing civil rights concerns and disinformation.

Schroepfer indicated that Facebook’s AI systems are designed to quickly adapt to changes in policy. “I don’t expect considerable technical changes are needed to adjust,” he told Spectrum.

This interview has been edited and condensed for clarity.

IEEE Spectrum: What are the stakes of content moderation? Is this an existential threat to Facebook? And is it critical that you deal well with the issue of election interference this year?

Schroepfer: It’s probably existential; it’s certainly massive. We are devoting a tremendous amount of our attention to it.

The idea that anyone could meddle in an election is deeply disturbing and offensive to all of us here, just as people and citizens of democracies. We don’t want to see that happen anywhere, and certainly not on our watch. So whether it’s important to the company or not, it’s important to us as people. And I feel a similar way on the content-moderation side.

There are not a lot of easy choices here. The only way to prevent people, with certainty, from posting bad things is to not let them post anything. We can take away all voice and just say, “Sorry, the Internet’s too dangerous. No one can use it.” That will certainly get rid of all hate speech online. But I don’t want to end up in that world. And there are variants of that world that various governments are trying to implement, where they get to decide what’s true or not, and you as a person don’t. I don’t want to get there either.

My hope is that we can build a set of tools that make it practical for us to do a good enough job, so that everyone is still excited about the idea that anyone can share what they want, and so that Facebook is a safe and reasonable place for people to operate in.

Spectrum: You joined Facebook in 2008, before AI was part of the company’s toolbox. When did that change? When did you begin to think that AI tools would be useful to Facebook?

Schroepfer: Ten years ago, AI wasn’t commercially practical; the technology just didn’t work very well. In 2012, there was one of those moments that a lot of people point to as the beginning of the current revolution in deep learning and AI. A computer-vision model—a neural network—was trained using what we call supervised training, and it turned out to be better than all the existing models.

Spectrum: How is that training done, and how did computer-vision models come to Facebook?

Image: Facebook

Just Broccoli? Facebook’s image analysis algorithms can tell the difference between marijuana [left] and tempura broccoli [right] better than some humans.

Schroepfer: Say I take a bunch of photos and I have people look at them. If they see a photo of a cat, they put a text label that says cat; if it’s one of a dog, the text label says dog. If you build a big enough data set and feed that to the neural net, it learns how to tell the difference between cats and dogs.

Prior to 2012, it didn’t work very well. And then in 2012, there was this moment where it seemed like, “Oh wow, this technique might work.” And a few years later we were deploying that form of technology to help us detect problematic imagery.

Spectrum: Do your AI systems work equally well on all types of prohibited content?

Schroepfer: Nudity was technically easiest. I don’t need to understand language or culture to understand that this is either a naked human or not. Violence is a much more nuanced problem, so it was harder technically to get it right. And with hate speech, not only do you have to understand the language, it may be very contextual, even tied to recent events. A week before the Christchurch shooting [New Zealand, 2019], saying “I wish you were in the mosque” probably doesn’t mean anything. A week after, that might be a terrible thing to say.

Spectrum: How much progress have you made on hate speech?

Schroepfer: AI, in the first quarter of 2020, proactively detected 88.8 percent of the hate-speech content we removed, up from 80.2 percent in the previous quarter. In the first quarter of 2020, we took action on 9.6 million pieces of content for violating our hate-speech policies.

Image: Facebook

Off Label: Sometimes image analysis isn’t enough to determine whether a picture posted violates the company’s policies. In considering these candy-colored vials of marijuana, for example, the algorithms can look at any accompanying text and, if necessary, comments on the post.

Spectrum: It sounds like you’ve expanded beyond tools that analyze images and are also using AI tools that analyze text.

Schroepfer: AI started off as very siloed. People worked on language, people worked on computer vision, people worked on video. We’ve put these things together—in production, not just as research—into multimodal classifiers.

[Schroepfer shows a photo of a pan of Rice Krispies treats, with text referring to it as a “potent batch”] This is a case in which you have an image, and then you have the text on the post. This looks like Rice Krispies. On its own, this image is fine. You put the text together with it in a bigger model; that can then understand what’s going on. That didn’t work five years ago.

Spectrum: Today, every post that goes up on Facebook is immediately checked by automated systems. Can you explain that process?

Image: Facebook

Bigger Picture: Identifying hate speech is often a matter of context. Either the text or the photo in this post isn’t hateful standing alone, but putting them together tells a different story.

Schroepfer: You upload an image and you write some text underneath it, and the systems look at both the image and the text to try to see which, if any, policies it violates. Those decisions are based on our Community Standards. It will also look at other signals on the posts, like the comments people make.

It happens relatively instantly, though there may be times things happen after the fact. Maybe you uploaded a post that had misinformation in it, and at the time you uploaded it, we didn’t know it was misinformation. The next day we fact-check something and scan again; we may find your post and take it down. As we learn new things, we’re going to go back through and look for violations of what we now know to be a problem. Or, as people comment on your post, we might update our understanding of it. If people are saying, “That’s terrible,” or “That’s mean,” or “That looks fake,” those comments may be an interesting signal.

Spectrum: How is Facebook applying its AI tools to the problem of election interference?

Schroepfer: I would split election interference into two categories. There are times when you’re going after the content, and there are times you’re going after the behavior or the authenticity of the person.

On content, if you’re sharing misinformation, saying, “It’s super Wednesday, not super Tuesday, come vote on Wednesday,” that’s a problem whether you’re an American sitting in California or a foreign actor.

Other times, people create a series of Facebook pages pretending they’re Americans, but they’re really a foreign entity. That is a problem on its own, even if all the content they’re sharing completely meets our Community Standards. The problem there is that you have a foreign government running an information operation.

There, you need different tools. What you’re trying to do is put pieces together, to say, “Wait a second. All of these pages—Martians for Justice, Moonlings for Justice, and Venusians for Justice”—are all run by an administrator with an IP address that’s outside the United States. So they’re all connected, even though they’re pretending to not be connected. That’s a very different problem than me sitting in my office in Menlo Park [Calif.] sharing misinformation.

I’m not going to go into lots of technical detail, because this is an area of adversarial nature. The fundamental problem you’re trying to solve is that there’s one entity coordinating the activity of a bunch of things that look like they’re not all one thing. So this is a series of Instagram accounts, or a series of Facebook pages, or a series of WhatsApp accounts, and they’re pretending to be totally different things. We’re looking for signals that these things are related in some way. And we’re looking through the graph [what Facebook calls its map of relationships between users] to understand the properties of this network.

Spectrum: What cutting-edge AI tools and methods have you been working on lately?

Schroepfer: Supervised learning, with humans setting up the instruction process for the AI systems, is amazingly effective. But it has a very obvious flaw: the speed at which you can develop these things is limited by how fast you can curate the data sets. If you’re dealing in a problem domain where things change rapidly, you have to rebuild a new data set and retrain the whole thing.

Self-supervision is inspired by the way people learn, by the way kids explore the world around them. To get computers to do it themselves, we take a bunch of raw data and build a way for the computer to construct its own tests. For language, you scan a bunch of Web pages, and the computer builds a test where it takes a sentence, eliminates one of the words, and figures out how to predict what word belongs there. And because it created the test, it actually knows the answer. I can use as much raw text as I can find and store because it’s processing everything itself and doesn’t require us to sit down and build the information set. In the last two years there has been a revolution in language understanding as a result of AI self-supervised learning.

Spectrum: What else are you excited about?

Schroepfer: What we’ve been working on over the last few years is multilingual understanding. Usually, when I’m trying to figure out, say, whether something is hate speech or not I have to go through the whole process of training the model in every language. I have to do that one time for every language. When you make a post, the first thing we have to figure out is what language your post is in. “Ah, that’s Spanish. So send it to the Spanish hate-speech model.”

We’ve started to build a multilingual model—one box where you can feed in text in 40 different languages and it determines whether it’s hate speech or not. This is way more effective and easier to deploy.

To geek out for a second, just the idea that you can build a model that understands a concept in multiple languages at once is crazy cool. And it not only works for hate speech, it works for a variety of things.

When we started working on this multilingual model years ago, it performed worse than every single individual model. Now, it not only works as well as the English model, but when you get to the languages where you don’t have enough data, it’s so much better. This rapid progress is very exciting.

Spectrum: How do you move new AI tools from your research labs into operational use?

Schroepfer: Engineers trying to make the next breakthrough will often say, “Cool, I’ve got a new thing and it achieved state-of-the-art results on machine translation.” And we say, “Great. How long does it take to run in production?” They say, “Well, it takes 10 seconds for every sentence to run on a CPU.” And we say, “It’ll eat our whole data center if we deploy that.” So we take that state-of-the-art model and we make it 10 or a hundred or a thousand times more efficient, maybe at the cost of a little bit of accuracy. So it’s not as good as the state-of-the-art version, but it’s something we can actually put into our data centers and run in production.

Spectrum: What’s the role of the humans in the loop? Is it true that Facebook currently employs 35,000 moderators?

Schroepfer: Yes. Right now our goal is not to reduce that. Our goal is to do a better job catching bad content. People often think that the end state will be a fully automated system. I don’t see that world coming anytime soon.

As automated systems get more sophisticated, they take more and more of the grunt work away, freeing up the humans to work on the really gnarly stuff where you have to spend an hour researching.

We also use AI to give our human moderators power tools. Say I spot this new meme that is telling everyone to vote on Wednesday rather than Tuesday. I have a tool in front of me that says, “Find variants of that throughout the system. Find every photo with the same text, find every video that mentions this thing and kill it in one shot.” Rather than, I found this one picture, but then a bunch of other people upload that misinformation in different forms.

Another important aspect of AI is that anything I can do to prevent a person from having to look at terrible things is time well spent. Whether it’s a person employed by us as a moderator or a user of our services, looking at these things is a terrible experience. If I can build systems that take the worst of the worst, the really graphic violence, and deal with that in an automated fashion, that’s worth a lot to me. Continue reading

Posted in Human Robots

#437749 Video Friday: NASA Launches Its Most ...

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):

AWS Cloud Robotics Summit – August 18-19, 2020 – [Virtual Conference]
CLAWAR 2020 – August 24-26, 2020 – [Virtual Conference]
ICUAS 2020 – September 1-4, 2020 – Athens, Greece
ICRES 2020 – September 28-29, 2020 – Taipei, Taiwan
AUVSI EXPONENTIAL 2020 – October 5-8, 2020 – [Virtual Conference]
IROS 2020 – October 25-29, 2020 – Las Vegas, Nevada
ICSR 2020 – November 14-16, 2020 – Golden, Colorado
Let us know if you have suggestions for next week, and enjoy today’s videos.

Yesterday was a big day for what was quite possibly the most expensive robot on Earth up until it wasn’t on Earth anymore.

Perseverance and the Ingenuity helicopter are expected to arrive on Mars early next year.

[ JPL ]

ICYMI, our most popular post this week featured Northeastern University roboticist John Peter Whitney literally putting his neck on the line for science! He was testing a remotely operated straight razor shaving robotic system powered by fluidic actuators. The cutting-edge (sorry!) device transmits forces from a primary stage, operated by a barber, to a secondary stage, with the razor attached.

[ John Peter Whitney ]

Together with Boston Dynamics, Ford is introducing a pilot program into our Van Dyke Transmission Plant. Say hello to Fluffy the Robot Dog, who creates fast and accurate 3D scans that helps Ford engineers when we’re retooling our plants.

Not shown in the video: “At times, Fluffy sits on its robotic haunches and rides on the back of a small, round Autonomous Mobile Robot, known informally as Scouter. Scouter glides smoothly up and down the aisles of the plant, allowing Fluffy to conserve battery power until it’s time to get to work. Scouter can autonomously navigate facilities while scanning and capturing 3-D point clouds to generate a CAD of the facility. If an area is too tight for Scouter, Fluffy comes to the rescue.”

[ Ford ]

There is a thing that happens at 0:28 in this video that I have questions about.

[ Ghost Robotics ]

Pepper is far more polite about touching than most humans.

[ Paper ]

We don’t usually post pure simulation videos unless they give us something to get really, really excited about. So here’s a pure simulation video.

[ Hybrid Robotics ]

University of Michigan researchers are developing new origami inspired methods for designing, fabricating and actuating micro-robots using heat.These improvements will expand the mechanical capabilities of the tiny bots, allowing them to fold into more complex shapes.

[ DRSL ]

HMI is making beastly electric arms work underwater, even if they’re not stapled to a robotic submarine.

[ HMI ]

Here’s some interesting work in progress from MIT’s Biomimetics Robotics Lab. The limb is acting as a “virtual magnet” using a bimodal force and direction sensor.

Thanks Peter!

[ MIT Biomimetics Lab ]

This is adorable but as a former rabbit custodian I can assure you that approximately 3 seconds after this video ended, all of the wires on that robot were chewed to bits.

[ Lingkang Zhang ]

During the ARCHE 2020 integration week, TNO and the ETH Robot System Lab (RSL) collaborated to integrate their research and development process using the Articulated Locomotion and MAnipulation (ALMA) robot. Next to the integration of software, we tested software to confirm proper implementation and development. We also captured visual and auditory data for future software development. This all resulted in the creation of multiple demo’s to show the capabilities of the teleoperation framework using the ALMA robot.

[ RSL ]

When we talk about practical applications quadrupedal robots with foot wheels, we don’t usually think about them on this scale, although we should.

[ RSL ]

Juan wrote in to share a DIY quadruped that he’s been working on, named CHAMP.

Juan says that the demo robot can be built in less than US $1000 with easily accessible parts. “I hope that my project can provide a more accessible platform for students, researchers, and enthusiasts who are interested to learn more about quadrupedal robot development and its underlying technology.”

[ CHAMP ]

Thanks Juan!

Here’s a New Zealand TV report about a study on robot abuse from Christoph Bartneck at the University of Canterbury.

[ Paper ]

Our Robotics Studio is a hands on class exposing students to practical aspects of the design, fabrication, and programming of physical robotic systems. So what happens when the class goes virtual due to the covid-19 virus? Things get physical — all @ home.

[ Columbia ]

A few videos from the Supernumerary Robotic Devices Workshop, held online earlier this month.

“Handheld Robots: Bridging the Gap between Fully External and Wearable Robots,” presented by Walterio Mayol-Cuevas, University of Bristol.

“Playing the Piano with 11 Fingers: The Neurobehavioural Constraints of Human Robot Augmentation,” presented by Aldo Faisal, Imperial College London.

[ Workshop ] Continue reading

Posted in Human Robots