Giving robots self-awareness as they move through space - and maybe even providing them with gene-like methods for storing rules of behavior - could be important steps toward creating more intelligent machines.
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.”
Basecamp Research is using portable labs like this one to gather samples from ecosystems around the world.
Clark works for Basecamp Research, a biotech company headquartered in London, and his job is to collect samples from ecosystems around the world. By extracting DNA from soil, water, plants, microbes and other organisms, Basecamp is building an extensive database of the Earth’s proteins. While DNA itself isn’t a protein, the information stored in DNA is used to create proteins, so extracting, sequencing, and annotating DNA allows for the discovery of unique protein sequences.
Using what they’re finding in the middle of the Atlantic and beyond, Basecamp’s detailed database is constantly growing. The outputs could be essential for cleaning up the damage done by toxic chemicals and finding alternatives to these chemicals.
Catalysts for change
Proteins provide structure and function in all living organisms. Some of these functional proteins are enzymes, which quite literally make things happen.
“Industrial chemistry is heavily polluting, especially the chemistry done in pharmaceutical drug development. Biocatalysis is providing advantages, both to make more complex drugs and to be more sustainable, reducing the pollution and toxicity of conventional chemistry," says Ahir Pushpanath, who heads partnerships for Basecamp.
“Enzymes are perfectly evolved catalysts,” says Ahir Pushpanath, a partnerships lead at Basecamp. ”Enzymes are essentially just a polymer, and polymers are made up of amino acids, which are nature’s building blocks.” He suggests thinking about it like Legos — if you have a bunch of Lego pieces and use them to build a structure that performs a function, “that’s basically how an enzyme works. In nature, these monuments have evolved to do life’s chemistry. If we didn’t have enzymes, we wouldn’t be alive.”
In our own bodies, enzymes catalyze everything from vision to digesting food to regrowing muscles, and these same types of enzymes are necessary in the pharmaceutical, agrochemical and fine chemical industries. But industrial conditions differ from those inside our bodies. So, when scientists need certain chemical reactions to create a particular product or substance, they make their own catalysts in their labs — generally through the use of petroleum and heavy metals.
These petrochemicals are effective and cost-efficient, but they’re wasteful and often hazardous. With growing concerns around sustainability and long-term public health, it's essential to find alternative solutions to toxic chemicals. “Industrial chemistry is heavily polluting, especially the chemistry done in pharmaceutical drug development,” Pushpanath says.
Basecamp is trying to replace lab-created catalysts with enzymes found in the wild. This concept is called biocatalysis, and in theory, all scientists have to do is find the right enzymes for their specific need. Yet, historically, researchers have struggled to find enzymes to replace petrochemicals. When they can’t identify a suitable match, they turn to what Pushpanath describes as “long, iterative, resource-intensive, directed evolution” in the laboratory to coax a protein into industrial adaptation. But the latest scientific advances have enabled these discoveries in nature instead.
Marlon Clark, a research scientist at Basecamp Research, looks for novel biochemistries in the Azores.
Glen Gowers
Enzyme hunters
Whether it’s Clark and a colleague setting off on an expedition, or a local, on-the-ground partner gathering and processing samples, there’s a lot to be learned from each collection. “Microbial genomes contain complete sets of information that define an organism — much like how letters are a code allowing us to form words, sentences, pages, and books that contain complex but digestible knowledge,” Clark says. He thinks of the environmental samples as biological libraries, filled with thousands of species, strains, and sequence variants. “It’s our job to glean genetic information from these samples.”
“We can actually dream up new proteins using generative AI," Pushpanath says.
Basecamp researchers manage this feat by sequencing the DNA and then assembling the information into a comprehensible structure. “We’re building the ‘stories’ of the biota,” Clark says. The more varied the samples, the more valuable insights his team gains into the characteristics of different organisms and their interactions with the environment. Sequencing allows scientists to examine the order of nucleotides — the organic molecules that form DNA — to identify genetic makeups and find changes within genomes. The process used to be too expensive, but the cost of sequencing has dropped from $10,000 a decade ago to as low as $100. Notably, biocatalysis isn’t a new concept — there have been waves of interest in using natural enzymes in catalysis for over a century, Pushpanath says. “But the technology just wasn’t there to make it cost effective,” he explains. “Sequencing has been the biggest boon.”
AI is probably the second biggest boon.
“We can actually dream up new proteins using generative AI,” Pushpanath says, which means that biocataylsis now has real potential to scale.
Glen Gowers, the co-founder of Basecamp, compares the company’s AI approach to that of social networks and streaming services. Consider how these platforms suggest connecting with the friends of your friends, or how watching one comedy film from the 1990s leads to a suggestion of three more.
“They’re thinking about data as networks of relationships as opposed to lists of items,” says Gowers. “By doing the same, we’re able to link the metadata of the proteins — by their relationships to each other, the environments in which they’re found, the way those proteins might look similar in sequence and structure, their surrounding genome context — really, this just comes down to creating a searchable network of proteins.”
On an Azores island, this volcanic opening may harbor organisms that can help scientists identify enzymes for biocatalysis to replace toxic chemicals.
Emma Bolton
Uwe Bornscheuer, professor at the Institute of Biochemistry at the University of Greifswald, and co-founder of Enzymicals, another biocatalysis company, says that the development of machine learning is a critical component of this work. “It’s a very hot topic, because the challenge in protein engineering is to predict which mutation at which position in the protein will make an enzyme suitable for certain applications,” Bornscheuer explains. These predictions are difficult for humans to make at all, let alone quickly. “It is clear that machine learning is a key technology.”
Benefiting from nature’s bounty
Biodiversity commonly refers to plants and animals, but the term extends to all life, including microbial life, and some regions of the world are more biodiverse than others. Building relationships with global partners is another key element to Basecamp’s success. Doing so in accordance with the access and benefit sharing principles set forth by the Nagoya Protocol — an international agreement that seeks to ensure the benefits of using genetic resources are distributed in a fair and equitable way — is part of the company's ethos. “There's a lot of potential for us, and there’s a lot of potential for our partners to have exactly the same impact in building and discovering commercially relevant proteins and biochemistries from nature,” Clark says.
Bornscheuer points out that Basecamp is not the first company of its kind. A former San Diego company called Diversa went public in 2000 with similar work. “At that time, the Nagoya Protocol was not around, but Diversa also wanted to ensure that if a certain enzyme or microorganism from Costa Rica, for example, were used in an industrial process, then people in Costa Rica would somehow profit from this.”
An eventual merger turned Diversa into Verenium Corporation, which is now a part of the chemical producer BASF, but it laid important groundwork for modern companies like Basecamp to continue to scale with today’s technologies.
“To collect natural diversity is the key to identifying new catalysts for use in new applications,” Bornscheuer says. “Natural diversity is immense, and over the past 20 years we have gained the advantages that sequencing is no longer a cost or time factor.”
This has allowed Basecamp to rapidly grow its database, outperforming Universal Protein Resource or UniProt, which is the public repository of protein sequences most commonly used by researchers. Basecamp’s database is three times larger, totaling about 900 million sequences. (UniProt isn’t compliant with the Nagoya Protocol, because, as a public database, it doesn’t provide traceability of protein sequences. Some scientists, however, argue that Nagoya compliance hinders progress.)
“Eventually, this work will reduce chemical processes. We’ll have cleaner processes, more sustainable processes," says Uwe Bornscheuer, a professor at the University of Greifswald.
With so much information available, Basecamp’s AI has been trained on “the true dictionary of protein sequence life,” Pushpanath says, which makes it possible to design sequences for particular applications. “Through deep learning approaches, we’re able to find protein sequences directly from our database, without the need for further laboratory-directed evolution.”
Recently, a major chemical company was searching for a specific transaminase — an enzyme that catalyzes a transfer of amino groups. “They had already spent a year-and-a-half and nearly two million dollars to evolve a public-database enzyme, and still had not reached their goal,” Pushpanath says. “We used our AI approaches on our novel database to yield 10 candidates within a week, which, when validated by the client, achieved the desired target even better than their best-evolved candidate.”
Basecamp’s other huge potential is in bioremediation, where natural enzymes can help to undo the damage caused by toxic chemicals. “Biocatalysis impacts both sides,” says Gowers. “It reduces the usage of chemicals to make products, and at the same time, where contamination sites do exist from chemical spills, enzymes are also there to kind of mop those up.”
So far, Basecamp's round-the-world sampling has covered 50 percent of the 14 major biomes, or regions of the planet that can be distinguished by their flora, fauna, and climate, as defined by the World Wildlife Fund. The other half remains to be catalogued — a key milestone for understanding our planet’s protein diversity, Pushpanath notes.
There’s still a long road ahead to fully replace petrochemicals with natural enzymes, but biocatalysis is on an upward trajectory. "Eventually, this work will reduce chemical processes,” Bornscheuer says. “We’ll have cleaner processes, more sustainable processes.”
In an initial study, Canary analyzed speech recordings with AI and identified early stage Alzheimer’s with 96 percent accuracy.
Arnold was diagnosed sooner than many others. It's important to find out early, when interventions can make the most difference. One promising avenue is looking at how people talk. Research has shown that Alzheimer’s affects a part of the brain that controls speech, resulting in small changes before people show other signs of the disease.
Now, Canary Speech, a company based in Utah, is using AI to examine elements like the pitch of a person’s voice and their pauses. In an initial study, Canary analyzed speech recordings with AI and identified early stage Alzheimer’s with 96 percent accuracy.
Developing the AI model
Canary Speech’s CEO, Henry O’Connell, met cofounder Jeff Adams about 40 years before they started the company. Back when they first crossed paths, they were both living in Bethesda, Maryland; O’Connell was a research fellow at the National Institutes of Health studying rare neurological diseases, while Adams was working to decode spy messages. Later on, Adams would specialize in building mathematical models to analyze speech and sound as a team leader in developing Amazon's Alexa.
It wasn't until 2015 that they decided to make use of the fit between their backgrounds. ““We established Canary Speech in 2017 to build a product that could be used in multiple languages in clinical environments,” O'Connell says.
The need is growing. About 55 million people worldwide currently live with Alzheimer’s, a number that is expected to double by 2050. Some scientists think the disease results from a buildup of plaque in the brain. It causes mild memory loss at first and, over time, this issue get worse while other symptoms, such as disorientation and hallucinations, can develop. Treatment to manage the disease is more effective in the earlier stages, but detection is difficult since mild symptoms are often attributed to the normal aging process.
O’Connell and Adams specialize in the complex ways that Alzheimer’s effects how people speak. Using AI, their mathematical model analyzes 15 million data points every minute, focusing on certain features of speech such as pitch, pauses and elongation of words. It also pays attention to how the vibrations of vocal cords change in different stages of the disease.
To create their model, the team used a type of machine learning called deep neural nets, which looks at multiple layers of data - in this case, the multiple features of a person’s speech patterns.
“Deep neural nets allow us to look at much, much larger data sets built out of millions of elements,” O’Connell explained. “Through machine learning and AI, we’ve identified features that are very sensitive to an Alzheimer’s patient versus [people without the disease] and also very sensitive to mild cognitive impairment, early stage and moderate Alzheimer's.” Based on their learnings, Canary is able to classify the disease stage very quickly, O’Connell said.
“When we’re listening to sublanguage elements, we’re really analyzing the direct result of changes in the brain in the physical body,” O’Connell said. “The brain controls your vocal cords: how fast they vibrate, the expansion of them, the contraction.” These factors, along with where people put their tongues when talking, function subconsciously and result in subtle changes in the sounds of speech.
Further testing is needed
In an initial trial, Canary analyzed speech recordings from phone calls to a large U.S. health insurer. They looked at the audio recordings of 651 policyholders who had early stage Alzheimer’s and 1018 who did not have the condition, aiming for a representative sample of age, gender and race. They used this data to create their first diagnostic model and found that it was 96 percent accurate in identifying Alzheimer’s.
Christian Herff, an assistant professor of neuroscience at Maastricht University in the Netherlands, praised this approach while adding that further testing is needed to assess its effectiveness.
“I think the general idea of identifying increased risk for cognitive impairment based on speech characteristics is very feasible, particularly when change in a user’s voice is monitored, for example, by recording speech every year,” Herff said. He noted that this can only be a first indication, not a full diagnosis. The accuracy still needs to be validated in studies that follows individuals over a period of time, he said.
Toby Walsh, a professor of artificial intelligence at the University of New South Wales, also thinks Canary’s tool has potential but highlights that Canary could diagnose some people who don’t really have the disease. “This is an interesting and promising application of AI,” he said, “but these tools need to be used carefully. Imagine the anxiety of being misdiagnosed with Alzheimer’s.”
As with many other AI tools, privacy and bias are additional issues to monitor closely, Walsh said.
Other languages
A related issue is that not everyone is fluent in English. Mahnaz Arvaneh, a senior lecturer in automatic control and systems engineering at the University of Sheffield, said this could be a blind spot.
“The system may not be very accurate for those who have English as their second language as their speaking patterns would be different, and any issue might be because of language deficiency rather than cognitive issues,” Arvaneh said.
The team is expanding to multiple languages starting with Japanese and Spanish. The elements of the model that make up the algorithm are very similar, but they need to be validated and retrained in a different language, which will require access to more data.
Recently, Canary analyzed the phone calls of 233 Japanese patients who had mild cognitive impairment and 704 healthy people. Using an English model they were able to identify the Japanese patients who had mild cognitive impairment with 78 percent accuracy. They also developed a model in Japanese that was 45 percent accurate, and they’re continuing to train it with more data.
The future
Canary is using their model to look at other diseases like Huntington’s and Parkinson’s. They’re also collaborating with pharmaceuticals to validate potential therapies for Alzheimer’s. By looking at speech patterns over time, Canary can get an indication of how well these drugs are working.
Dave Arnold and his wife dance at his nephew’s wedding in Rochester, New York, ten years ago, before his Alzheimer's diagnosis.
Dave Arnold
Ultimately, they want to integrate their tool into everyday life. “We want it to be used in a smartphone, or a teleconference call so that individuals could be examined in their home,” O’Connell said. “We could follow them over time and work with clinical teams and hospitals to improve the evaluation of patients and contribute towards an accurate diagnosis.”
Arnold, the patient with early stage Alzheimer’s, sees great promise. “The process of getting a diagnosis is already filled with so much anxiety,” he said. “Anything that can be done to make it easier and less stressful would be a good thing, as long as it’s proven accurate.”