A newly published study by University of Michigan researchers shows facial recognition technology in schools presents multiple problems and has limited efficacy. Led by Shobita Parthasarathy, director of the university's Science, Technology, and Public Policy (STPP) program, the research say the technology isn't suited to security purposes and can actively promote racial discrimination, normalize surveillance, and erode privacy while institutionalizing inaccuracy and marginalizing non-conforming students. The study follows the New York legislature's passage of a moratorium on the use of facial recognition and other forms of biometric identification in schools until 2022. The bill, which came in response to the launch of facial recognition by the Lockport City School District, was among the first in the nation to explicitly regulate or ban use of the technology in schools. That development came after companies including Amazon, IBM, and Microsoft halted or ended the sale of facial recognition products in response to the first wave of Black Lives Matter protests in the U.S. The Michigan University study -- a part of STPP's Technology Assessment Project -- employs an analogical case comparison method to look at previous uses of security technology like CCTV cameras and metal detectors as well as biometric technologies and anticipate the implications of facial recognition.
Would you entrust a personal-injury claim, divorce settlement or high-stakes contract to an algorithm? A growing number of apps and digital services are betting you will, attracting millions of Silicon Valley investment dollars but raising questions about the limits and ethics of technology in the legal sphere. Among the leaders in the emergent robo-lawyering field is DoNotPay, an app dreamed up by Joshua Browder in 2015, when he was a 17-year-old Stanford University student, to help friends dispute parking tickets. The app, which relies on an artificial intelligence-enabled chatbot, became popular, and has expanded its focus to other consumer legal services. In June it hit the million-case mark, helping save people upward of $30 million since it started, Mr. Browder says. It raised a new $12 million round of funding the same month.
The U.S. Department of Education's Institute of Education Sciences has awarded the National Center for Research on Education Access and Choice (REACH) at Tulane University a $100,000 contract to collect data from approximately 150,000 school websites across the country to see how the nation's education system is responding to the coronavirus pandemic. The project, which will track traditional public schools, charter schools and private schools, aims to quickly answer questions that are critical for understanding how students are learning when school buildings are closed. Key questions include: how many schools are providing any kind of instructional support; which are delivering online instruction; what resources are they offering to students and how do students stay in contact with teachers? "This data will also help answer important questions about equity in the school system, showing how responses differ according to characteristics like spending levels, student demographics, internet access, and if there are differences based on whether it is a private, charter or traditional public school," said REACH National Director Douglas N. Harris, Schlieder Foundation Chair in Public Education and chair of economics at Tulane University School of Liberal Arts. REACH will work in cooperation with Nicholas Mattei, assistant professor of computer science at Tulane University School of Science and Engineering, to create a computer program that will collect data from every school and district website in the country.
Every time Congress holds a hearing about Silicon Valley companies, people mock the legislators for being out of their depth. Last week's effort by the antitrust subcommittee of the House Judiciary Committee was no exception. "The technological ignorance demonstrated by our elected officials ... was truly stunning," Shelly Palmer, CEO at the Palmer Group, a tech strategy advisory group, told USA Today. "People who are this clueless about the economic forces shaping our world should not be tasked with leading us into the age of AI," he said. "The data elite are playing a different game with a different set of rules. Apparently, Congress can't even find the ballpark."
The state said it has no formal reporting process for tracking coronavirus outbreaks that have already cropped up in summer school programs, leaving teachers unions wondering how health officials plan to prevent outbreaks considered "inevitable" in the fall. "We are not formally tracking them, but we are trying to notice them as they pop up," said Department of Elementary and Secondary Education spokeswoman Jacqueline Reis. "There is no formal reporting process for schools." Reis said the DESE is still finalizing its guidance as schools shore up their plans for remote, in-person or hybrid learning once classes resume in September. "It's absurd and it's stunning but its also not a surprise," said Merrie Najimy, who leads the Massachusetts Teachers Association.
Analyttica TreasureHunt LEAPS is a Data Science Learning & Practice platform. It's innovative application focused courses & hands-on learning techniques, enable users to build their own models with pre-loaded datasets. Our programs facilitate interactions with industry expert & mentors. Get certified & stay ahead of the competition. Learn Data Science & Machine Learning from Zero to Hero in 29 Weeks: Big Bang Data Science Solutions has developed an ecosystem training program to boost your Data Science career to the next level.
Nine philosophers explore the various issues and questions raised by the newly released language model, GPT-3, in this edition of Philosophers On, guest edited by Annette Zimmermann. Introduction Annette Zimmermann, guest editor GPT-3, a powerful, 175 billion parameter language model developed recently by OpenAI, has been galvanizing public debate and controversy. As the MIT Technology Review puts it: “OpenAI’s new language generator GPT-3 is shockingly good—and completely mindless”. Parts of the technology community hope (and fear) that GPT-3 could brings us one step closer to the hypothetical future possibility of human-like, highly sophisticated artificial general intelligence (AGI). Meanwhile, others (including OpenAI’s own CEO) have critiqued claims about GPT-3’s ostensible proximity to AGI, arguing that they are vastly overstated. Why the hype? As is turns out, GPT-3 is unlike other natural language processing (NLP) systems, the latter of which often struggle with what comes comparatively easily to humans: performing entirely new language tasks based on a few simple instructions and examples. Instead, NLP systems usually have to be pre-trained on a large corpus of text, and then fine-tuned in order to successfully perform a specific task. GPT-3, by contrast, does not require fine tuning of this kind: it seems to be able to perform a whole range of tasks reasonably well, from producing fiction, poetry, and press releases to functioning code, and from music, jokes, and technical manuals, to “news articles which human evaluators have difficulty distinguishing from articles written by humans”. The Philosophers On series contains group posts on issues of current interest, with the aim being to show what the careful thinking characteristic of philosophers (and occasionally scholars in related fields) can bring to popular ongoing conversations. Contributors present not fully worked out position papers but rather brief thoughts that can serve as prompts for further reflection and discussion. The contributors to this installment of “Philosophers On” are Amanda Askell (Research Scientist, OpenAI), David Chalmers (Professor of Philosophy, New York University), Justin Khoo (Associate Professor of Philosophy, Massachusetts Institute of Technology), Carlos Montemayor (Professor of Philosophy, San Francisco State University), C. Thi Nguyen (Associate Professor of Philosophy, University of Utah), Regina Rini (Canada Research Chair in Philosophy of Moral and Social Cognition, York University), Henry Shevlin (Research Associate, Leverhulme Centre for..
For Jonathan Losos, tiny Caribbean islands and their reptile inhabitants are test tubes of evolution. The morning of 17 October 1996 started as usual for Jonathan Losos. The evolutionary biologist donned a broad hat and slathered on sunscreen, then headed by boat to several unnamed islets off Great Exuma Island in the Bahamas. Three years earlier, he and ecologist David Spiller had introduced local lizard species there to learn how they would compete in a once lizardless place. The pair spent the day snaring lizards, noting their exact locations, and taking stock of the insects, spiders, and vegetation. They were worried about reports of an impending hurricane, but the locals seemed confident it would veer off and spare the islands, as usual. Not this time, however. The next day, Losos and Spiller helped their hotel owner board up the windows of their beachfront cottage on Great Exuma as Hurricane Lili bore down on the island. As the wind picked up and the first squalls dumped rain, they scurried to a cinder block building up a hill. That night, the wind blew off parts of the roof and felled palm trees. A 4-meter storm surge flooded the streets, and 2 days later they found their rented motorboat stuck in a tree. The lizards had it even worse. When Losos and Spiller finally made it back out to their most exposed study sites, the islands were stripped nearly bare of brush and all the lizards were gone. But the setback for Losos's project was the start of a new chapter in his research on how the animals adapt to the varied, changeable environments on islands in and around the Caribbean. Since Lili, a half-dozen other hurricanes have inundated islets and swept away animals relocated there by Losos, who is based at Washington University in St. Louis (WashU), and his team. But he and his colleagues have persevered, collecting data on how the animals adapt to predators, storm damage, and other challenges—natural and those contrived by the researchers. A lifelong reptile enthusiast, Losos is driven in part by his passion for a group of lizards called anoles, which thrive in South and Central America and throughout the Caribbean. He also views them as an opportunity. Almost half of the 400 anole species live on islands, and their diverse lifestyles, habitats, and histories have proved to be a vehicle for exploring some of evolution's biggest questions. “Jonathan's islands are like giant test tubes, and he is the ultimate tinkerer,” says Martha Muñoz, an evolutionary biologist at Yale University. Losos's research on anoles has shown that evolution can happen faster than most scientists had assumed, and that—contrary to what some leading thinkers have proposed—it is often predictable. Faced with similar challenges, separate populations often evolve similar solutions. Along the way, Losos has mentored dozens of young scientists, and some are now carrying his work in new directions. “Beyond his many contributions to the field, Jonathan has also changed the course of science simply by being who he is,” Muñoz, a former student, says. “He is proof that success is richer and more rewarding when accompanied by kindness and humility.” ODDLY ENOUGH, THE 1950S TV show Leave it to Beaver started Losos down this path. When 7-year-old Beaver brought home a pet alligator, young Losos asked his parents whether he, too, could get one. His mom was against it, but his father said he would ask a family friend, the deputy director of the St. Louis Zoo, for advice. A successful businessman, the senior Losos also loved animals, taking his family on nature vacations, joining the zoo's board, and even financing the zoo's acquisition of a baby elephant from Thailand, which he named Carolyn in honor of his wife. To everyone's surprise, the director heartily approved, saying that having an alligator as a childhood pet was how he got his start in herpetology. So the junior Losos acquired several baby caimans, which lived in a baby pool in the basement in winter and in a horse trough in the yard the rest of the year. Only a few times did the animals escape and terrorize the neighbors. Losos worked summers at the zoo until partway through college, eventually donating his caimans to a zookeeper. “Jonathan started off as a little kid loving nature, endlessly pestering staff at his local zoo, catching lizards on family vacations, and he's never lost that spark,” says Harry Greene, herpetologist emeritus at Cornell University and Losos's graduate school adviser. As an undergraduate at Harvard University, Losos fell under the tutelage of herpetologist Ernest Williams. Sometimes referred to as the father of anole biology, Williams had recognized that anoles on different Caribbean islands evolved independently. Yet on each island he'd found a similar set of body types or “ecomorphs”—one specialized for living in the brush, another for gripping twigs, and still others for life high in the trees. These parallels suggested that where circumstances were similar, evolution would converge on the same set of traits and form communities with similar sets of species. Williams's lab had already produced several leading evolutionary biologists, and Losos figured the field of anole research was getting too crowded. But no other species both captured his interest and was easy to study. “I went through a dozen failed Ph.D. projects,” he recalls. At a low point, he seriously considered law school, but his dad convinced him that the world needed herpetologists more than lawyers. Losos eventually realized that anoles were perfect for applying new tools in evolutionary biology. Researchers were just beginning to build family trees and trace evolution based on protein variations among species. For his Ph.D., Losos compared proteins in Caribbean anoles and verified that Williams's ecomorphs had indeed evolved independently to form similar communities on different islands ( Science , 27 March 1998, p. ). That insight alone—support for an idea called convergent evolution—“was a really important breakthrough,” says Frank Burbrink, a herpetologist at the American Museum of Natural History. Meanwhile, other researchers were calling for more rigor in evolution studies by requiring evidence that supposedly adaptive traits really give an organism an advantage. So Losos began to study different anole ecomorphs, with legs and toepads of varying sizes (see graphic, p. 499). In the lab, he ran them down miniature racetracks and assessed how well they clung to smooth, vertical surfaces. He found that lizards living near the ground, close to predators, had longer legs that made them fast, whereas those living higher in brush and trees had bigger toepads to stick to leaves and smooth bark. By combining these data with his family tree studies, he got a clearer sense of the lizards' evolutionary history. He “was really one of the first people to move the field into doing evolution by integrating ecology and morphology and getting the bigger picture,” Burbrink says. Inspired by experiments in which researchers monitored evolutionary changes in guppies in Trinidad after relocating them to different streams ( Science , 24 August 2012, p. ), Losos began to wonder whether similar studies could be done in Caribbean anoles. And he realized that Thomas Schoener, one of Williams's protégés, had already laid the groundwork. In the 1980s, Thomas and Amy Schoener (they were once married) introduced local lizards to tiny lizardless islands in the Bahamas to investigate how different vegetation affected the reptiles' ability to thrive. A decade later, Losos teamed up with Thomas Schoener, by then a renowned ecologist at the University of California (UC), Davis, to revisit those sites. Consistent with Williams's and Losos's earlier findings, lizards living in scrubby vegetation had shorter legs and larger toepads than their ancestors, which had lived in tall, broad trees. These adaptations enabled them to cling to tiny twigs as they chased down insects to eat, and the changes had taken just a few generations. “Evolution can happen very quickly when natural selection is very strong,” Losos says. The idea is now well-accepted, but at the time it went against the entrenched belief that evolution was a slow process. “This is one of the few things that [Charles] Darwin got wrong,” Losos says. He decided to make anoles his life's work. HE SOON HAD TO RECKON with hurricanes. Losos and Spiller, now retired from UC Davis, had chosen the islets off Great Exuma to study the effects of competition. On some, they introduced two local species, the green and brown anoles, and on others, just a single species. In the first 3 years, they noticed that on islands with both kinds, the brown lizards were driving the green anoles higher into the bushes, where they were struggling. That's when Lili hit, ruining the experiment before they could see whether the green anoles would go extinct. “It would have been so easy, I'm sure, to pack it all in and give up,” says Luke Harmon, one of Losos's former students and now an evolutionary biologist at the University of Idaho. Instead, Losos and Spiller used the disaster to their advantage. They documented Lili's great, but also patchy, impact. Islands southwest of Great Exuma felt the brunt of the storm surge and were devoid of lizards and vegetation. Life there would have to start over. On islands to the north, the wind and rain snapped twigs and ripped off leaves but a few lizards remained, they reported in the first of several papers about hurricanes ( Science , 31 July 1998, p. ). The work challenged a widespread assumption that extreme events such as hurricanes do not drive evolution because they are rare and have random, unpredictable impacts on plants and animals. The Losos group discovered instead that storms can be agents of natural selection. For example, in 2017 Losos's postdoc Colin Donihue; functional morphologist Anthony Herrel, now with the French national research agency CNRS at the National Museum of Natural History; and colleagues visited two cays in the Turks and Caicos to measure the body proportions of the anoles living there. Four days after they left, two almost back-to-back hurricanes hit the area with winds of more than 200 kilometers per hour. When the team returned a few weeks later and remeasured the lizards, they found that the survivors tended to have bigger toepads, longer forelimbs, and shorter hindlimbs. Back in the lab, the researchers tested how these traits affect the lizards' ability to hold onto a perch. In a strong wind, anoles hang on with their forelimbs, but they lose their grip with the hind legs. Cranking up an air-blower, the researchers found that those with longer hind legs (and more surface area for the wind to catch) got blown off their perches onto a padded surface more readily. Conversely, animals with shorter hind limbs and bigger toepads hung on. The hurricanes had apparently selected for those traits, the team reported in 2018. The following year, they found that offspring of the survivors also had big toepads, suggesting the adaptation was genetic and not just a reaction to holding on tight. The team has since measured toepad size in 188 lizard species across the Caribbean. The more hurricanes an island has experienced, the bigger the toepads of the lizards living there, they reported on 27 April in the Proceedings of the National Academy of Sciences . Hurricanes seem to have had a long-term evolutionary effect. LOSOS HAD BEEN A PROFESSOR at WashU for 13 years when Harvard came calling in 2005, seeking to recruit him to its evolutionary biology department. A St. Louis native and a hardcore St. Louis Cardinals baseball fan, he hesitated. He even did a yearlong sabbatical at Harvard before finally accepting, in large part because the position included a curatorship at the Harvard Museum of Comparative Zoology. “That was the one thing St. Louis didn't have,” he recalls. There, he continued to build on a reputation for being a kind, enthusiastic mentor. “I have seen him give high school students the same attention and respect that he gives his closest colleagues,” says Melissa Kemp, a former postdoc now at the University of Texas, Austin. “He seems to always be focused on his work, but he also has a whimsical sense of fun at the same time,” says Michele Johnson, a former student and an evolutionary biologist at Trinity University. Losos sports a watch with an anole he photographed as its face and is not above lecturing undergraduates while dressed as a platypus—one of his favorite animals since childhood. Those traits and a firm belief that “there is no one-size-fits-all in terms of how to interact with and mentor students” have helped Losos launch the careers of 59 graduate students and postdocs. They include at least eight Black, Latino, and Native American scholars, in a field that lacks diversity. (Although 3% of U.S. biologists are African American or Black, for example, only 0.3% of evolutionary biologists are.) Ambika Kamath, now a postdoctoral researcher at UC Berkeley, says Losos backed her completely when her studies challenged the long-held idea that male lizards hold territories to corral their mates. She argued instead that females move around and play a role in mate choice. “It would have been much harder for me to do that work without his excitement,” she recalls. Losos worked hard with her to get the paper just right and was eager to be a co-author. “Otherwise it would have just been the work of this young brown woman who could have easily been dismissed as an angry feminist.” Kamath and other students praise Losos for pushing them intellectually without undermining their confidence. Harmon jokes that Losos would never dismiss an idea from his students, no matter how wacky. Instead, he would just pause and say “interesting.” “Eventually I figured out that maybe I should think things through a bit more, if Jonathan thought they were ‘interesting,’” Harmon says. LOSOS AND HIS TEAM keep testing their ideas about ecology and evolution on Caribbean islands. In one recent project, Robert Pringle, now at Princeton University, and Losos tested a key principle in ecology—that introducing a top predator tends to increase biodiversity. The researchers added a predatory ground-dwelling lizard to islands with brown and green anoles. To escape this new threat, the brown anoles began to hang out higher in the foliage, displacing the green anoles that normally lived there and driving them toward extinction. Contrary to conventional wisdom, the predator appeared to be pushing the islands toward lower biodiversity, they reported on 5 June in Nature . Another recent study, led by one of Losos's former postdocs, examined the impacts of an invasive anole species on Dominica. Until 20 years ago, the island was home to a single anole species. Then a lumber shipment introduced a second species that is gradually spreading. To study how the native and invader species interact, behavioral ecologist Claire Dufour, now at the University of Montpellier, used robotic lizards as stand-ins for the invader. The robots did pushups and extended a flap of fake skin under the chin, mimicking the aggressive displays of real lizards. In response, native lizards familiar with the invaders postured more aggressively, suggesting the invaders are forcing the natives to expend more energy defending their territory, the group reported on 27 March in the Journal of Animal Ecology . “Our biggest conclusion is that the species do compete and have negative consequences on each other,” Losos says. ![Figure] Evolution's stamp on island-dwelling lizards On islands in and around the Caribbean, 173 species of anole lizards face an array of different environments, predators, and competitors, along with periodic storms. The result is a laboratory of evolution, where scientists have been able to track the speed and course of adaptation. GRAPHIC: V. ALTOUNIAN/ SCIENCE Even as his group continues to churn out papers, Losos is assessing what he has learned so far. In his book Improbable Destinies: Fate, Chance, and the Future of Evolution , published in 2017, he challenged a major contention of one of the field's great thinkers, Stephen Jay Gould, the Harvard paleontologist who argued that chance plays such a big role in determining nature's course that evolution would never take the same path twice. Anoles offer evidence to the contrary, Losos wrote: In similar habitats, they have repeatedly evolved similar body shapes, sizes, and behavior. The book was written for the general public, but it made an impression even on his peers. “I've been studying evolution for 30-plus years, and this book made me rethink some things I thought I knew about biology and evolution,” says Christopher Austin, an evolutionary biologist and herpetology curator at the Louisiana State University Museum of Natural Science. Losos left Harvard in 2018, lured by a new job at WashU and the prospect of returning to his hometown, his cats, and his wife, who has a successful real estate career and did not follow him to Massachusetts. He now heads the Living Earth Collaborative, a biodiversity research initiative that unites experts at the Missouri Botanical Garden, the St. Louis Zoo, and WashU. He is working on a book about evolution in the house cat, another of his favorite species. And he is still dodging hurricanes. Losos and colleagues have been trying to assess the long-term evolutionary impacts of predatory lizards they've introduced to some islands in the Bahamas. “I don't know if we will ever get there,” he says. Every few years a hurricane comes through and blows the evolving lizards away. : http://www.sciencemag.org/content/279/5359/2115 : http://www.sciencemag.org/content/337/6097/904 : http://www.sciencemag.org/content/281/5377/695 : pending:yes
Amy McGovern is a Lloyd G. and Joyce Austin Presidential Professor in the School of Computer Science and an Adjunct Professor in the School of Meteorology at the University of Oklahoma. She has been leading the development of AI/ML for weather applications for 15 years. As climate change affects weather patterns and sea levels rise, the world's need for accurate, usable predictions of weather and ocean and their impacts has never been greater. At the same time, the quantity and quality of Earth observation and modeling systems are increasing dramatically, offering a deluge of data so rich that only automated intelligent systems can fully exploit it. In this talk, I will discuss our approach to developing trustworthy AI methods for environmental science.