Tag Archives: recognize

#435626 Video Friday: Watch Robots Make a Crepe ...

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. Every week, we also post a calendar of upcoming robotics events; here's what we have so far (send us your events!):

Robotronica – August 18, 2019 – Brisbane, Australia
CLAWAR 2019 – August 26-28, 2019 – Kuala Lumpur, Malaysia
IEEE Africon 2019 – September 25-27, 2019 – Accra, Ghana
ISRR 2019 – October 6-10, 2019 – Hanoi, Vietnam
Ro-Man 2019 – October 14-18, 2019 – New Delhi
Humanoids 2019 – October 15-17, 2019 – Toronto
ARSO 2019 – October 31-November 2, 2019 – Beijing
ROSCon 2019 – October 31-November 1, 2019 – Macau
IROS 2019 – November 4-8, 2019 – Macau
Let us know if you have suggestions for next week, and enjoy today's videos.

Team CoSTAR (JPL, MIT, Caltech, KAIST, LTU) has one of the more diverse teams of robots that we’ve seen:

[ Team CoSTAR ]

A team from Carnegie Mellon University and Oregon State University is sending ground and aerial autonomous robots into a Pittsburgh-area mine to prepare for this month’s DARPA Subterranean Challenge.

“Look at that fire extinguisher, what a beauty!” Expect to hear a lot more of that kind of weirdness during SubT.

[ CMU ]

Unitree Robotics is starting to batch-manufacture Laikago Pro quadrupeds, and if you buy four of them, they can carry you around in a chair!

I’m also really liking these videos from companies that are like, “We have a whole bunch of robot dogs now—what weird stuff can we do with them?”

[ Unitree Robotics ]

Why take a handful of pills every day for all the stuff that's wrong with you, when you could take one custom pill instead? Because custom pills are time-consuming to make, that’s why. But robots don’t care!

Multiply Labs’ factory is designed to operate in parallel. All the filling robots and all the quality-control robots are operating at the same time. The robotic arm, in the meanwhile, shuttles dozens of trays up and down the production floor, making sure that each capsule is filled with the right drugs. The manufacturing cell shown in this article can produce 10,000 personalized capsules in an 8-hour shift. A single cell occupies just 128 square feet (12 square meters) on the production floor. This means that a regular production facility (~10,000 square feet, or 929 m2 ) can house 78 cells, for an overall output of 780,000 capsules per shift. This exceeds the output of most traditional manufacturers—while producing unique personalized capsules!

[ Multiply Labs ]

Thanks Fred!

If you’re getting tired of all those annoying drones that sound like giant bees, just have a listen to this turbine-powered one:

[ Malloy Aeronautics ]

In retrospect, it’s kind of amazing that nobody has bothered to put a functional robotic dog head on a quadruped robot before this, right?

Equipped with sensors, high-tech radar imaging, cameras and a directional microphone, this 100-pound (45-kilogram) super-robot is still a “puppy-in-training.” Just like a regular dog, he responds to commands such as “sit,” “stand,” and “lie down.” Eventually, he will be able to understand and respond to hand signals, detect different colors, comprehend many languages, coordinate his efforts with drones, distinguish human faces, and even recognize other dogs.

As an information scout, Astro’s key missions will include detecting guns, explosives and gun residue to assist police, the military, and security personnel. This robodog’s talents won’t just end there, he also can be programmed to assist as a service dog for the visually impaired or to provide medical diagnostic monitoring. The MPCR team also is training Astro to serve as a first responder for search-and-rescue missions such as hurricane reconnaissance as well as military maneuvers.

[ FAU ]

And now this amazing video, “The Coke Thief,” from ICRA 2005 (!):

[ Paper ]

CYBATHLON Series put the focus on one or two of the six disciplines and are organized in cooperation with international universities and partners. The CYBATHLON Arm and Leg Prosthesis Series took place in Karlsruhe, Germany, from 16 to 18 May and was organized in cooperation with the Karlsruhe Institute of Technology (KIT) and the trade fair REHAB Karlsruhe.

The CYBATHLON Wheelchair Series took place in Kawasaki, Japan on 5 May 2019 and was organized in cooperation with the CYBATHLON Wheelchair Series Japan Organizing Committee and supported by the Swiss Embassy.

[ Cybathlon ]

Rainbow crepe robot!

There’s also this other robot, which I assume does something besides what's in the video, because otherwise it appears to be a massively overengineered way of shaping cooked rice into a chubby triangle.

[ PC Watch ]

The Weaponized Plastic Fighting League at Fetch Robotics has had another season of shardation, deintegration, explodification, and other -tions. Here are a couple fan favorite match videos:

[ Fetch Robotics ]

This video is in German, but it’s worth watching for the three seconds of extremely satisfying footage showing a robot twisting dough into pretzels.

[ Festo ]

Putting brains into farming equipment is a no-brainer, since it’s a semi-structured environment that's generally clear of wayward humans driving other vehicles.

[ Lovol ]

Thanks Fan!

Watch some robots assemble suspiciously Lego-like (but definitely not actually Lego) minifigs.

[ DevLinks ]

The Robotics Innovation Facility (RIFBristol) helps businesses, entrepreneurs, researchers and public sector bodies to embrace the concept of ‘Industry 4.0'. From training your staff in robotics, and demonstrating how automation can improve your manufacturing processes, to prototyping and validating your new innovations—we can provide the support you need.

[ RIF ]

Ryan Gariepy from Clearpath Robotics (and a bunch of other stuff) gave a talk at ICRA with the title of “Move Fast and (Don’t) Break Things: Commercializing Robotics at the Speed of Venture Capital,” which is more interesting when you know that this year’s theme was “Notable Failures.”

[ Clearpath Robotics ]

In this week’s episode of Robots in Depth, Per interviews Michael Nielsen, a computer vision researcher at the Danish Technological Institute.

Michael worked with a fusion of sensors like stereo vision, thermography, radar, lidar and high-frame-rate cameras, merging multiple images for high dynamic range. All this, to be able to navigate the tricky situation in a farm field where you need to navigate close to or even in what is grown. Multibaseline cameras were also used to provide range detection over a wide range of distances.

We also learn about how he expanded his work into sorting recycling, a very challenging problem. We also hear about the problems faced when using time of flight and sheet of light cameras. He then shares some good results using stereo vision, especially combined with blue light random dot projectors.

[ Robots in Depth ] Continue reading

Posted in Human Robots

#435614 3 Easy Ways to Evaluate AI Claims

When every other tech startup claims to use artificial intelligence, it can be tough to figure out if an AI service or product works as advertised. In the midst of the AI “gold rush,” how can you separate the nuggets from the fool’s gold?

There’s no shortage of cautionary tales involving overhyped AI claims. And applying AI technologies to health care, education, and law enforcement mean that getting it wrong can have real consequences for society—not just for investors who bet on the wrong unicorn.

So IEEE Spectrum asked experts to share their tips for how to identify AI hype in press releases, news articles, research papers, and IPO filings.

“It can be tricky, because I think the people who are out there selling the AI hype—selling this AI snake oil—are getting more sophisticated over time,” says Tim Hwang, director of the Harvard-MIT Ethics and Governance of AI Initiative.

The term “AI” is perhaps most frequently used to describe machine learning algorithms (and deep learning algorithms, which require even less human guidance) that analyze huge amounts of data and make predictions based on patterns that humans might miss. These popular forms of AI are mostly suited to specialized tasks, such as automatically recognizing certain objects within photos. For that reason, they are sometimes described as “weak” or “narrow” AI.

Some researchers and thought leaders like to talk about the idea of “artificial general intelligence” or “strong AI” that has human-level capacity and flexibility to handle many diverse intellectual tasks. But for now, this type of AI remains firmly in the realm of science fiction and is far from being realized in the real world.

“AI has no well-defined meaning and many so-called AI companies are simply trying to take advantage of the buzz around that term,” says Arvind Narayanan, a computer scientist at Princeton University. “Companies have even been caught claiming to use AI when, in fact, the task is done by human workers.”

Here are three ways to recognize AI hype.

Look for Buzzwords
One red flag is what Hwang calls the “hype salad.” This means stringing together the term “AI” with many other tech buzzwords such as “blockchain” or “Internet of Things.” That doesn’t automatically disqualify the technology, but spotting a high volume of buzzwords in a post, pitch, or presentation should raise questions about what exactly the company or individual has developed.

Other experts agree that strings of buzzwords can be a red flag. That’s especially true if the buzzwords are never really explained in technical detail, and are simply tossed around as vague, poorly-defined terms, says Marzyeh Ghassemi, a computer scientist and biomedical engineer at the University of Toronto in Canada.

“I think that if it looks like a Google search—picture ‘interpretable blockchain AI deep learning medicine’—it's probably not high-quality work,” Ghassemi says.

Hwang also suggests mentally replacing all mentions of “AI” in an article with the term “magical fairy dust.” It’s a way of seeing whether an individual or organization is treating the technology like magic. If so—that’s another good reason to ask more questions about what exactly the AI technology involves.

And even the visual imagery used to illustrate AI claims can indicate that an individual or organization is overselling the technology.

“I think that a lot of the people who work on machine learning on a day-to-day basis are pretty humble about the technology, because they’re largely confronted with how frequently it just breaks and doesn't work,” Hwang says. “And so I think that if you see a company or someone representing AI as a Terminator head, or a big glowing HAL eye or something like that, I think it’s also worth asking some questions.”

Interrogate the Data

It can be hard to evaluate AI claims without any relevant expertise, says Ghassemi at the University of Toronto. Even experts need to know the technical details of the AI algorithm in question and have some access to the training data that shaped the AI model’s predictions. Still, savvy readers with some basic knowledge of applied statistics can search for red flags.

To start, readers can look for possible bias in training data based on small sample sizes or a skewed population that fails to reflect the broader population, Ghassemi says. After all, an AI model trained only on health data from white men would not necessarily achieve similar results for other populations of patients.

“For me, a red flag is not demonstrating deep knowledge of how your labels are defined.”
—Marzyeh Ghassemi, University of Toronto

How machine learning and deep learning models perform also depends on how well humans labeled the sample datasets use to train these programs. This task can be straightforward when labeling photos of cats versus dogs, but gets more complicated when assigning disease diagnoses to certain patient cases.

Medical experts frequently disagree with each other on diagnoses—which is why many patients seek a second opinion. Not surprisingly, this ambiguity can also affect the diagnostic labels that experts assign in training datasets. “For me, a red flag is not demonstrating deep knowledge of how your labels are defined,” Ghassemi says.

Such training data can also reflect the cultural stereotypes and biases of the humans who labeled the data, says Narayanan at Princeton University. Like Ghassemi, he recommends taking a hard look at exactly what the AI has learned: “A good way to start critically evaluating AI claims is by asking questions about the training data.”

Another red flag is presenting an AI system’s performance through a single accuracy figure without much explanation, Narayanan says. Claiming that an AI model achieves “99 percent” accuracy doesn’t mean much without knowing the baseline for comparison—such as whether other systems have already achieved 99 percent accuracy—or how well that accuracy holds up in situations beyond the training dataset.

Narayanan also emphasized the need to ask questions about an AI model’s false positive rate—the rate of making wrong predictions about the presence of a given condition. Even if the false positive rate of a hypothetical AI service is just one percent, that could have major consequences if that service ends up screening millions of people for cancer.

Readers can also consider whether using AI in a given situation offers any meaningful improvement compared to traditional statistical methods, says Clayton Aldern, a data scientist and journalist who serves as managing director for Caldern LLC. He gave the hypothetical example of a “super-duper-fancy deep learning model” that achieves a prediction accuracy of 89 percent, compared to a “little polynomial regression model” that achieves 86 percent on the same dataset.

“We're talking about a three-percentage-point increase on something that you learned about in Algebra 1,” Aldern says. “So is it worth the hype?”

Don’t Ignore the Drawbacks

The hype surrounding AI isn’t just about the technical merits of services and products driven by machine learning. Overblown claims about the beneficial impacts of AI technology—or vague promises to address ethical issues related to deploying it—should also raise red flags.

“If a company promises to use its tech ethically, it is important to question if its business model aligns with that promise,” Narayanan says. “Even if employees have noble intentions, it is unrealistic to expect the company as a whole to resist financial imperatives.”

One example might be a company with a business model that depends on leveraging customers’ personal data. Such companies “tend to make empty promises when it comes to privacy,” Narayanan says. And, if companies hire workers to produce training data, it’s also worth asking whether the companies treat those workers ethically.

The transparency—or lack thereof—about any AI claim can also be telling. A company or research group can minimize concerns by publishing technical claims in peer-reviewed journals or allowing credible third parties to evaluate their AI without giving away big intellectual property secrets, Narayanan says. Excessive secrecy is a big red flag.

With these strategies, you don’t need to be a computer engineer or data scientist to start thinking critically about AI claims. And, Narayanan says, the world needs many people from different backgrounds for societies to fully consider the real-world implications of AI.

Editor’s Note: The original version of this story misspelled Clayton Aldern’s last name as Alderton. Continue reading

Posted in Human Robots

#435593 AI at the Speed of Light

Neural networks shine for solving tough problems such as facial and voice recognition, but conventional electronic versions are limited in speed and hungry for power. In theory, optics could beat digital electronic computers in the matrix calculations used in neural networks. However, optics had been limited by their inability to do some complex calculations that had required electronics. Now new experiments show that all-optical neural networks can tackle those problems.

The key attraction of neural networks is their massive interconnections among processors, comparable to the complex interconnections among neurons in the brain. This lets them perform many operations simultaneously, like the human brain does when looking at faces or listening to speech, making them more efficient for facial and voice recognition than traditional electronic computers that execute one instruction at a time.

Today's electronic neural networks have reached eight million neurons, but their future use in artificial intelligence may be limited by their high power usage and limited parallelism in connections. Optical connections through lenses are inherently parallel. The lens in your eye simultaneously focuses light from across your field of view onto the retina in the back of your eye, where an array of light-detecting nerve cells detects the light. Each cell then relays the signal it receives to neurons in the brain that process the visual signals to show us an image.

Glass lenses process optical signals by focusing light, which performs a complex mathematical operation called a Fourier transform that preserves the information in the original scene but rearranges is completely. One use of Fourier transforms is converting time variations in signal intensity into a plot of the frequencies present in the signal. The military used this trick in the 1950s to convert raw radar return signals recorded by an aircraft in flight into a three-dimensional image of the landscape viewed by the plane. Today that conversion is done electronically, but the vacuum-tube computers of the 1950s were not up to the task.

Development of neural networks for artificial intelligence started with electronics, but their AI applications have been limited by their slow processing and need for extensive computing resources. Some researchers have developed hybrid neural networks, in which optics perform simple linear operations, but electronics perform more complex nonlinear calculations. Now two groups have demonstrated simple all-optical neural networks that do all processing with light.

In May, Wolfram Pernice of the Institute of Physics at the University of Münster in Germany and colleagues reported testing an all-optical “neuron” in which signals change target materials between liquid and solid states, an effect that has been used for optical data storage. They demonstrated nonlinear processing, and produced output pulses like those from organic neurons. They then produced an integrated photonic circuit that incorporated four optical neurons operating at different wavelengths, each of which connected to 15 optical synapses. The photonic circuit contained more than 140 components and could recognize simple optical patterns. The group wrote that their device is scalable, and that the technology promises “access to the high speed and high bandwidth inherent to optical systems, thus enabling the direct processing of optical telecommunication and visual data.”

Now a group at the Hong Kong University of Science and Technology reports in Optica that they have made an all-optical neural network based on a different process, electromagnetically induced transparency, in which incident light affects how atoms shift between quantum-mechanical energy levels. The process is nonlinear and can be triggered by very weak light signals, says Shengwang Du, a physics professor and coauthor of the paper.

In their demonstration, they illuminated rubidium-85 atoms cooled by lasers to about 10 microKelvin (10 microdegrees above absolute zero). Although the technique may seem unusually complex, Du said the system was the most accessible one in the lab that could produce the desired effects. “As a pure quantum atomic system [it] is ideal for this proof-of-principle experiment,” he says.

Next, they plan to scale up the demonstration using a hot atomic vapor center, which is less expensive, does not require time-consuming preparation of cold atoms, and can be integrated with photonic chips. Du says the major challenges are reducing cost of the nonlinear processing medium and increasing the scale of the all-optical neural network for more complex tasks.

“Their demonstration seems valid,” says Volker Sorger, an electrical engineer at George Washington University in Washington who was not involved in either demonstration. He says the all-optical approach is attractive because it offers very high parallelism, but the update rate is limited to about 100 hertz because of the liquid crystals used in their test, and he is not completely convinced their approach can be scaled error-free. Continue reading

Posted in Human Robots

#434865 5 AI Breakthroughs We’ll Likely See in ...

Convergence is accelerating disruption… everywhere! Exponential technologies are colliding into each other, reinventing products, services, and industries.

As AI algorithms such as Siri and Alexa can process your voice and output helpful responses, other AIs like Face++ can recognize faces. And yet others create art from scribbles, or even diagnose medical conditions.

Let’s dive into AI and convergence.

Top 5 Predictions for AI Breakthroughs (2019-2024)
My friend Neil Jacobstein is my ‘go-to expert’ in AI, with over 25 years of technical consulting experience in the field. Currently the AI and Robotics chair at Singularity University, Jacobstein is also a Distinguished Visiting Scholar in Stanford’s MediaX Program, a Henry Crown Fellow, an Aspen Institute moderator, and serves on the National Academy of Sciences Earth and Life Studies Committee. Neil predicted five trends he expects to emerge over the next five years, by 2024.

AI gives rise to new non-human pattern recognition and intelligence results

AlphaGo Zero, a machine learning computer program trained to play the complex game of Go, defeated the Go world champion in 2016 by 100 games to zero. But instead of learning from human play, AlphaGo Zero trained by playing against itself—a method known as reinforcement learning.

Building its own knowledge from scratch, AlphaGo Zero demonstrates a novel form of creativity, free of human bias. Even more groundbreaking, this type of AI pattern recognition allows machines to accumulate thousands of years of knowledge in a matter of hours.

While these systems can’t answer the question “What is orange juice?” or compete with the intelligence of a fifth grader, they are growing more and more strategically complex, merging with other forms of narrow artificial intelligence. Within the next five years, who knows what successors of AlphaGo Zero will emerge, augmenting both your business functions and day-to-day life.

Doctors risk malpractice when not using machine learning for diagnosis and treatment planning

A group of Chinese and American researchers recently created an AI system that diagnoses common childhood illnesses, ranging from the flu to meningitis. Trained on electronic health records compiled from 1.3 million outpatient visits of almost 600,000 patients, the AI program produced diagnosis outcomes with unprecedented accuracy.

While the US health system does not tout the same level of accessible universal health data as some Chinese systems, we’ve made progress in implementing AI in medical diagnosis. Dr. Kang Zhang, chief of ophthalmic genetics at the University of California, San Diego, created his own system that detects signs of diabetic blindness, relying on both text and medical images.

With an eye to the future, Jacobstein has predicted that “we will soon see an inflection point where doctors will feel it’s a risk to not use machine learning and AI in their everyday practices because they don’t want to be called out for missing an important diagnostic signal.”

Quantum advantage will massively accelerate drug design and testing

Researchers estimate that there are 1060 possible drug-like molecules—more than the number of atoms in our solar system. But today, chemists must make drug predictions based on properties influenced by molecular structure, then synthesize numerous variants to test their hypotheses.

Quantum computing could transform this time-consuming, highly costly process into an efficient, not to mention life-changing, drug discovery protocol.

“Quantum computing is going to have a major industrial impact… not by breaking encryption,” said Jacobstein, “but by making inroads into design through massive parallel processing that can exploit superposition and quantum interference and entanglement, and that can wildly outperform classical computing.”

AI accelerates security systems’ vulnerability and defense

With the incorporation of AI into almost every aspect of our lives, cyberattacks have grown increasingly threatening. “Deep attacks” can use AI-generated content to avoid both human and AI controls.

Previous examples include fake videos of former President Obama speaking fabricated sentences, and an adversarial AI fooling another algorithm into categorizing a stop sign as a 45 mph speed limit sign. Without the appropriate protections, AI systems can be manipulated to conduct any number of destructive objectives, whether ruining reputations or diverting autonomous vehicles.

Jacobstein’s take: “We all have security systems on our buildings, in our homes, around the healthcare system, and in air traffic control, financial organizations, the military, and intelligence communities. But we all know that these systems have been hacked periodically and we’re going to see that accelerate. So, there are major business opportunities there and there are major opportunities for you to get ahead of that curve before it bites you.”

AI design systems drive breakthroughs in atomically precise manufacturing

Just as the modern computer transformed our relationship with bits and information, AI will redefine and revolutionize our relationship with molecules and materials. AI is currently being used to discover new materials for clean-tech innovations, such as solar panels, batteries, and devices that can now conduct artificial photosynthesis.

Today, it takes about 15 to 20 years to create a single new material, according to industry experts. But as AI design systems skyrocket in capacity, these will vastly accelerate the materials discovery process, allowing us to address pressing issues like climate change at record rates. Companies like Kebotix are already on their way to streamlining the creation of chemistries and materials at the click of a button.

Atomically precise manufacturing will enable us to produce the previously unimaginable.

Final Thoughts
Within just the past three years, countries across the globe have signed into existence national AI strategies and plans for ramping up innovation. Businesses and think tanks have leaped onto the scene, hiring AI engineers and tech consultants to leverage what computer scientist Andrew Ng has even called the new ‘electricity’ of the 21st century.

As AI plays an exceedingly vital role in everyday life, how will your business leverage it to keep up and build forward?

In the wake of burgeoning markets, new ventures will quickly arise, each taking advantage of untapped data sources or unmet security needs.

And as your company aims to ride the wave of AI’s exponential growth, consider the following pointers to leverage AI and disrupt yourself before it reaches you first:

Determine where and how you can begin collecting critical data to inform your AI algorithms
Identify time-intensive processes that can be automated and accelerated within your company
Discern which global challenges can be expedited by hyper-fast, all-knowing minds

Remember: good data is vital fuel. Well-defined problems are the best compass. And the time to start implementing AI is now.

Join Me
Abundance-Digital Online Community: I’ve created a Digital/Online community of bold, abundance-minded entrepreneurs called Abundance-Digital. Abundance-Digital is my ‘onramp’ for exponential entrepreneurs – those who want to get involved and play at a higher level. Click here to learn more.

Image Credit: Yurchanka Siarhei / 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