Scientists Want to Make Robots with Genomes that Help Grow their Minds
One day in recent past, scientists at Columbia University’s Creative Machines Lab set up a robotic arm inside a circle of five streaming video cameras and let the robot watch itself move, turn and twist. For about three hours the robot did exactly that—it looked at itself this way and that, like toddlers exploring themselves in a room full of mirrors. By the time the robot stopped, its internal neural network finished learning the relationship between the robot’s motor actions and the volume it occupied in its environment. In other words, the robot built a spatial self-awareness, just like humans do. “We trained its deep neural network to understand how it moved in space,” says Boyuan Chen, one of the scientists who worked on it.
For decades robots have been doing helpful tasks that are too hard, too dangerous, or physically impossible for humans to carry out themselves. Robots are ultimately superior to humans in complex calculations, following rules to a tee and repeating the same steps perfectly. But even the biggest successes for human-robot collaborations—those in manufacturing and automotive industries—still require separating the two for safety reasons. Hardwired for a limited set of tasks, industrial robots don't have the intelligence to know where their robo-parts are in space, how fast they’re moving and when they can endanger a human.
Over the past decade or so, humans have begun to expect more from robots. Engineers have been building smarter versions that can avoid obstacles, follow voice commands, respond to human speech and make simple decisions. Some of them proved invaluable in many natural and man-made disasters like earthquakes, forest fires, nuclear accidents and chemical spills. These disaster recovery robots helped clean up dangerous chemicals, looked for survivors in crumbled buildings, and ventured into radioactive areas to assess damage.
Now roboticists are going a step further, training their creations to do even better: understand their own image in space and interact with humans like humans do. Today, there are already robot-teachers like KeeKo, robot-pets like Moffin, robot-babysitters like iPal, and robotic companions for the elderly like Pepper.
But even these reasonably intelligent creations still have huge limitations, some scientists think. “There are niche applications for the current generations of robots,” says professor Anthony Zador at Cold Spring Harbor Laboratory—but they are not “generalists” who can do varied tasks all on their own, as they mostly lack the abilities to improvise, make decisions based on a multitude of facts or emotions, and adjust to rapidly changing circumstances. “We don’t have general purpose robots that can interact with the world. We’re ages away from that.”
Robotic spatial self-awareness – the achievement by the team at Columbia – is an important step toward creating more intelligent machines. Hod Lipson, professor of mechanical engineering who runs the Columbia lab, says that future robots will need this ability to assist humans better. Knowing how you look and where in space your parts are, decreases the need for human oversight. It also helps the robot to detect and compensate for damage and keep up with its own wear-and-tear. And it allows robots to realize when something is wrong with them or their parts. “We want our robots to learn and continue to grow their minds and bodies on their own,” Chen says. That’s what Zador wants too—and on a much grander level. “I want a robot who can drive my car, take my dog for a walk and have a conversation with me.”
Columbia scientists have trained a robot to become aware of its own "body," so it can map the right path to touch a ball without running into an obstacle, in this case a square.
Jane Nisselson and Yinuo Qin/ Columbia Engineering
Today’s technological advances are making some of these leaps of progress possible. One of them is the so-called Deep Learning—a method that trains artificial intelligence systems to learn and use information similar to how humans do it. Described as a machine learning method based on neural network architectures with multiple layers of processing units, Deep Learning has been used to successfully teach machines to recognize images, understand speech and even write text.
Trained by Google, one of these language machine learning geniuses, BERT, can finish sentences. Another one called GPT3, designed by San Francisco-based company OpenAI, can write little stories. Yet, both of them still make funny mistakes in their linguistic exercises that even a child wouldn’t. According to a paper published by Stanford’s Center for Research on Foundational Models, BERT seems to not understand the word “not.” When asked to fill in the word after “A robin is a __” it correctly answers “bird.” But try inserting the word “not” into that sentence (“A robin is not a __”) and BERT still completes it the same way. Similarly, in one of its stories, GPT3 wrote that if you mix a spoonful of grape juice into your cranberry juice and drink the concoction, you die. It seems that robots, and artificial intelligence systems in general, are still missing some rudimentary facts of life that humans and animals grasp naturally and effortlessly.
How does one give robots a genome? Zador has an idea. We can’t really equip machines with real biological nucleotide-based genes, but we can mimic the neuronal blueprint those genes create.
It's not exactly the robots’ fault. Compared to humans, and all other organisms that have been around for thousands or millions of years, robots are very new. They are missing out on eons of evolutionary data-building. Animals and humans are born with the ability to do certain things because they are pre-wired in them. Flies know how to fly, fish knows how to swim, cats know how to meow, and babies know how to cry. Yet, flies don’t really learn to fly, fish doesn’t learn to swim, cats don’t learn to meow, and babies don’t learn to cry—they are born able to execute such behaviors because they’re preprogrammed to do so. All that happens thanks to the millions of years of evolutions wired into their respective genomes, which give rise to the brain’s neural networks responsible for these behaviors. Robots are the newbies, missing out on that trove of information, Zador argues.
A neuroscience professor who studies how brain circuitry generates various behaviors, Zador has a different approach to developing the robotic mind. Until their creators figure out a way to imbue the bots with that information, robots will remain quite limited in their abilities. Each model will only be able to do certain things it was programmed to do, but it will never go above and beyond its original code. So Zador argues that we have to start giving robots a genome.
How does one do that? Zador has an idea. We can’t really equip machines with real biological nucleotide-based genes, but we can mimic the neuronal blueprint those genes create. Genomes lay out rules for brain development. Specifically, the genome encodes blueprints for wiring up our nervous system—the details of which neurons are connected, the strength of those connections and other specs that will later hold the information learned throughout life. “Our genomes serve as blueprints for building our nervous system and these blueprints give rise to a human brain, which contains about 100 billion neurons,” Zador says.
If you think what a genome is, he explains, it is essentially a very compact and compressed form of information storage. Conceptually, genomes are similar to CliffsNotes and other study guides. When students read these short summaries, they know about what happened in a book, without actually reading that book. And that’s how we should be designing the next generation of robots if we ever want them to act like humans, Zador says. “We should give them a set of behavioral CliffsNotes, which they can then unwrap into brain-like structures.” Robots that have such brain-like structures will acquire a set of basic rules to generate basic behaviors and use them to learn more complex ones.
Currently Zador is in the process of developing algorithms that function like simple rules that generate such behaviors. “My algorithms would write these CliffsNotes, outlining how to solve a particular problem,” he explains. “And then, the neural networks will use these CliffsNotes to figure out which ones are useful and use them in their behaviors.” That’s how all living beings operate. They use the pre-programmed info from their genetics to adapt to their changing environments and learn what’s necessary to survive and thrive in these settings.
For example, a robot’s neural network could draw from CliffsNotes with “genetic” instructions for how to be aware of its own body or learn to adjust its movements. And other, different sets of CliffsNotes may imbue it with the basics of physical safety or the fundamentals of speech.
At the moment, Zador is working on algorithms that are trying to mimic neuronal blueprints for very simple organisms—such as earthworms, which have only 302 neurons and about 7000 synapses compared to the millions we have. That’s how evolution worked, too—expanding the brains from simple creatures to more complex to the Homo Sapiens. But if it took millions of years to arrive at modern humans, how long would it take scientists to forge a robot with human intelligence? That’s a billion-dollar question. Yet, Zador is optimistic. “My hypotheses is that if you can build simple organisms that can interact with the world, then the higher level functions will not be nearly as challenging as they currently are.”
This is part 2 of a three part series on a new generation of doctors leading the charge to make the health care industry more sustainable - for the benefit of their patients and the planet. Read part 1 here.
After graduating from her studies as an engineer, Nora Stroetzel ticked off the top item on her bucket list and traveled the world for a year. She loved remote places like the Indonesian rain forest she reached only by hiking for several days on foot, mountain villages in the Himalayas, and diving at reefs that were only accessible by local fishing boats.
“But no matter how far from civilization I ventured, one thing was already there: plastic,” Stroetzel says. “Plastic that would stay there for centuries, on 12,000 foot peaks and on beaches several hundred miles from the nearest city.” She saw “wild orangutans that could be lured by rustling plastic and hermit crabs that used plastic lids as dwellings instead of shells.”
While traveling she started volunteering for beach cleanups and helped build a recycling station in Indonesia. But the pivotal moment for her came after she returned to her hometown Kiel in Germany. “At the dentist, they gave me a plastic cup to rinse my mouth. I used it for maybe ten seconds before it was tossed out,” Stroetzel says. “That made me really angry.”
She decided to research alternatives for plastic in the medical sector and learned that cups could be reused and easily disinfected. All dentists routinely disinfect their tools anyway and, Stroetzel reasoned, it wouldn’t be too hard to extend that practice to cups.
It's a good example for how often plastic is used unnecessarily in medical practice, she says. The health care sector is the fifth biggest source of pollution and trash in industrialized countries. In the U.S., hospitals generate an estimated 6,000 tons of waste per day, including an average of 400 grams of plastic per patient per day, and this sector produces 8.5 percent of greenhouse gas emissions nationwide.
“Sustainable alternatives exist,” Stroetzel says, “but you have to painstakingly look for them; they are often not offered by the big manufacturers, and all of this takes way too much time [that] medical staff simply does not have during their hectic days.”
When Stroetzel spoke with medical staff in Germany, she found they were often frustrated by all of this waste, especially as they took care to avoid single-use plastic at home. Doctors in other countries share this frustration. In a recent poll, nine out of ten doctors in Germany said they’re aware of the urgency to find sustainable solutions in the health industry but don’t know how to achieve this goal.
After a year of researching more sustainable alternatives, Stroetzel founded a social enterprise startup called POP, short for Practice Without Plastic, together with IT expert Nicolai Niethe, to offer well-researched solutions. “Sustainable alternatives exist,” she says, “but you have to painstakingly look for them; they are often not offered by the big manufacturers, and all of this takes way too much time [that] medical staff simply does not have during their hectic days.”
In addition to reusable dentist cups, other good options for the heath care sector include washable N95 face masks and gloves made from nitrile, which waste less water and energy in their production. But Stroetzel admits that truly making a medical facility more sustainable is a complex task. “This includes negotiating with manufacturers who often package medical materials in double and triple layers of extra plastic.”
While initiatives such as Stroetzel’s provide much needed information, other experts reason that a wholesale rethinking of healthcare is needed. Voluntary action won’t be enough, and government should set the right example. Kari Nadeau, a Stanford physician who has spent 30 years researching the effects of environmental pollution on the immune system, and Kenneth Kizer, the former undersecretary for health in the U.S. Department of Veterans Affairs, wrote in JAMA last year that the medical industry and federal agencies that provide health care should be required to measure and make public their carbon footprints. “Government health systems do not disclose these data (and very rarely do private health care organizations), unlike more than 90% of the Standard & Poor’s top 500 companies and many nongovernment entities," they explained. "This could constitute a substantial step toward better equipping health professionals to confront climate change and other planetary health problems.”
Compared to the U.K., the U.S. healthcare industry lags behind in terms of measuring and managing its carbon footprint, and hospitals are the second highest energy user of any sector in the U.S.
Kizer and Nadeau look to the U.K. National Health Service (NHS), which created a Sustainable Development Unit in 2008 and began that year to conduct assessments of the NHS’s carbon footprint. The NHS also identified its biggest culprits: Of the 2019 footprint, with emissions totaling 25 megatons of carbon dioxide equivalent, 62 percent came from the supply chain, 24 percent from the direct delivery of care, 10 percent from staff commute and patient and visitor travel, and 4 percent from private health and care services commissioned by the NHS. From 1990 to 2019, the NHS has reduced its emission of carbon dioxide equivalents by 26 percent, mostly due to the switch to renewable energy for heat and power. Meanwhile, the NHS has encouraged health clinics in the U.K. to install wind generators or photovoltaics that convert light to electricity -- relatively quick ways to decarbonize buildings in the health sector.
Compared to the U.K., the U.S. healthcare industry lags behind in terms of measuring and managing its carbon footprint, and hospitals are the second highest energy user of any sector in the U.S. “We are already seeing patients with symptoms from climate change, such as worsened respiratory symptoms from increased wildfires and poor air quality in California,” write Thomas B. Newman, a pediatrist at the University of California, San Francisco, and UCSF clinical research coordinator Daisy Valdivieso. “Because of the enormous health threat posed by climate change, health professionals should mobilize support for climate mitigation and adaptation efforts.” They believe “the most direct place to start is to approach the low-lying fruit: reducing healthcare waste and overuse.”
In addition to resulting in waste, the plastic in hospitals ultimately harms patients, who may be even more vulnerable to the effects due to their health conditions. Microplastics have been detected in most humans, and on average, a human ingests five grams of microplastic per week. Newman and Valdivieso refer to the American Board of Internal Medicine's Choosing Wisely program as one of many initiatives that identify and publicize options for “safely doing less” as a strategy to reduce unnecessary healthcare practices, and in turn, reduce cost, resource use, and ultimately reduce medical harm.
A few U.S. clinics are pioneers in transitioning to clean energy sources. In Wisconsin, the nonprofit Gundersen Health network became the first hospital to cut its reliance on petroleum by switching to locally produced green energy in 2015, and it saved $1.2 million per year in the process. Kaiser Permanente eliminated its 800,000 ton carbon footprint through energy efficiency and purchasing carbon offsets, reaching a balance between carbon emissions and removing carbon from the atmosphere in 2020, the first U.S. health system to do so.
Cleveland Clinic has pledged to join Kaiser in becoming carbon neutral by 2027. Realizing that 80 percent of its 2008 carbon emissions came from electricity consumption, the Clinic started switching to renewable energy and installing solar panels, and it has invested in researching recyclable products and packaging. The Clinic’s sustainability report outlines several strategies for producing less waste, such as reusing cases for sterilizing instruments, cutting back on materials that can’t be recycled, and putting pressure on vendors to reduce product packaging.
The Charité Berlin, Europe’s biggest university hospital, has also announced its goal to become carbon neutral. Its sustainability managers have begun to identify the biggest carbon culprits in its operations. “We’ve already reduced CO2 emissions by 21 percent since 2016,” says Simon Batt-Nauerz, the director of infrastructure and sustainability.
The hospital still emits 100,000 tons of CO2 every year, as much as a city with 10,000 residents, but it’s making progress through ride share and bicycle programs for its staff of 20,000 employees, who can get their bikes repaired for free in one of the Charité-operated bike workshops. Another program targets doctors’ and nurses’ scrubs, which cause more than 200 tons of CO2 during manufacturing and cleaning. The staff is currently testing lighter, more sustainable scrubs made from recycled cellulose that is grown regionally and requires 80 percent less land use and 30 percent less water.
The Charité hospital in Berlin still emits 100,000 tons of CO2 every year, but it’s making progress through ride share and bicycle programs for its staff of 20,000 employees.
Anesthesiologist Susanne Koch spearheads sustainability efforts in anesthesiology at the Charité. She says that up to a third of hospital waste comes from surgery rooms. To reduce medical waste, she recommends what she calls the 5 Rs: Reduce, Reuse, Recycle, Rethink, Research. “In medicine, people don’t question the use of plastic because of safety concerns,” she says. “Nobody wants to be sued because something is reused. However, it is possible to reduce plastic and other materials safely.”
For instance, she says, typical surgery kits are single-use and contain more supplies than are actually needed, and the entire kit is routinely thrown out after the surgery. “Up to 20 percent of materials in a surgery room aren’t used but will be discarded,” Koch says. One solution could be smaller kits, she explains, and another would be to recycle the plastic. Another example is breathing tubes. “When they became scarce during the pandemic, studies showed that they can be used seven days instead of 24 hours without increased bacteria load when we change the filters regularly,” Koch says, and wonders, “What else can we reuse?”
In the Netherlands, TU Delft researchers Tim Horeman and Bart van Straten designed a method to melt down the blue polypropylene wrapping paper that keeps medical instruments sterile, so that the material can be turned it into new medical devices. Currently, more than a million kilos of the blue paper are used in Dutch hospitals every year. A growing number of Dutch hospitals are adopting this approach.
Another common practice that’s ripe for improvement is the use of a certain plastic, called PVC, in hospital equipment such as blood bags, tubes and masks. Because of its toxic components, PVC is almost never recycled in the U.S., but University of Michigan researchers Danielle Fagnani and Anne McNeil have discovered a chemical process that can break it down into material that could be incorporated back into production. This could be a step toward a circular economy “that accounts for resource inputs and emissions throughout a product’s life cycle, including extraction of raw materials, manufacturing, transport, use and reuse, and disposal,” as medical experts have proposed. “It’s a failure of humanity to have created these amazing materials which have improved our lives in many ways, but at the same time to be so shortsighted that we didn’t think about what to do with the waste,” McNeil said in a press release.
Susanne Koch puts it more succinctly: “What’s the point if we save patients while killing the planet?”
The Friday Five covers five stories in research that you may have missed this week. There are plenty of controversies and troubling ethical issues in science – and we get into many of them in our online magazine – but this news roundup focuses on scientific creativity and progress to give you a therapeutic dose of inspiration headed into the weekend.
Here are the promising studies covered in this week's Friday Five:
- Kids stressing you out? They could be protecting your health.
- A new device unlocks the heart's secrets
- Super-ager gene transplants
- Surgeons could 3D print your organs before operations
- A skull cap looks into the brain like an fMRI