We are pleased to kick-off the Computational Sustainability Virtual Seminar Series with a talk by Professor Stefano Ermon of Stanford University, co-author on ”Combining satellite imagery and machine learning to predict poverty” (Science, August 19, 2016) .
Anyone may register here to receive connection details on this Zoom webinar (it’s free!). Please distribute this link to others who may be interested, especially colleagues and students.
The Computational Sustainability Virtual Seminar Series first talk is scheduled for
Tuesday, September 27, 2016
4:00 – 5:00 pm Eastern Time (1:00 – 2:00 pm Pacific Time)
Measuring progress towards sustainable development goals with machine learning Stefano Ermon, Stanford University
Recent technological developments are creating new spatio-temporal data streams that contain a wealth of information relevant to sustainable development goals. Modern AI techniques have the potential to yield accurate, inexpensive, and highly scalable models to inform research and policy. As a first example, I will present a machine learning method we developed to predict and map poverty in developing countries. Our method can reliably predict economic well-being using only high-resolution satellite imagery. Because images are passively collected in every corner of the world, our method can provide timely and accurate measurements in a very scalable end economic way, and could revolutionize efforts towards global poverty eradication. As a second example, I will present some ongoing work on monitoring agricultural and food security outcomes from space.
BIO: Stefano Ermon is currently an Assistant Professor in the Department of Computer Science at Stanford University, where he is affiliated with the Artificial Intelligence Laboratory and the Woods Institute for the Environment. He completed his PhD in computer science at Cornell in 2015. His research interests include techniques for scalable and accurate inference in graphical models, statistical modeling of data, large-scale combinatorial optimization, and robust decision making under uncertainty, and is motivated by a range of applications, in particular ones in the emerging field of computational sustainability. Stefano has won several awards, including two Best Student Paper Awards, one Runner-Up Prize, and a McMullen Fellowship.
The Computational Sustainability Virtual Seminar Series will present talks by researchers and educators in Computational Sustainability, and is being sponsored by CompSustNet, with support from the National Science Foundation’s Expeditions in Computing program.
Carla Gomes and Doug Fisher
Watch this video to learn more about the Ermon Lab’s “Combining Satellite Imagery and Machine Learning to predict poverty”
I first met Andrew when he was a first year Ph.D. candidate in Economics and he dropped by my office to chat about his interests in development, particularly in Africa. We had many stimulating conversations and I wound up supervising his dissertation, which ranged from applied microeconomic theory applied to questions of microfinance and informal lending for education – a common practice in rural Kenya – to empirical work on the functioning of farmers cooperatives based on original survey data he collected in rural Kenya. Toward the end of his doctoral studies, Andrew worked as my research assistant on a project using data from the Kenyan government’s early warning system in the pastoral drylands. As we studied those data we uncovered systematic patterns that led us to believe that one might be able to predict herd losses statistically with some precision.
Another of my Ph.D. students, Sommarat Chantarat (Economics PhD 2009) then took up the challenge of designing an index-based livestock insurance (IBLI) contract in her dissertation, which won national and international awards. Dr. Chantarat, Dr. Mude – who moved to the Nairobi-based International Livestock Research Institute (ILRI) after completing his Cornell degree — and I worked closely with other partners to get IBLI designed and commercially piloted beginning in early 2010. We have had an extensive collaborative work agreement with Dr. Mude’s team at ILRI since 2008, which has supported Dr. Chantarat’s work as well as a more recent Ph.D., Nathan Jensen (AEM, 2014) whose dissertation evaluating the impacts and uptake of IBLI also won international recognition. This has been an especially enjoyable, fruitful, and impactful collaboration, which we are delighted continues quite actively today. The range of projects we have undertaken together has broadened over time, increasingly including computationally intensive work with Institute for Computational Sustainability faculty and staff.
This work to help address systemic risk among some of the poorest and most marginalized populations in the world is tremendously important and exciting. It is an enormous privilege to work with a collaborator as talented, committed and kind as Dr. Mude.
Opinion originally published in The Hill on August 26, 2016 By Amanda D. Rodewald
With political divisiveness so often headlining the news, how refreshing it is to celebrate a centennial that demonstrates the power of countries coming together. One hundred years ago, the United States and Great Britain (for Canada) came together for birds when they signed the Migratory Bird Treaty, a convention to protect migratory birds across international borders.
At the time, populations of many birds were plummeting due to poorly regulated hunting. The plume trade, for which an estimated 5 million birds — especially waterbirds like egrets and herons — were killed each year for feathers to adorn hats, eventually incited people to action. In response, the landmark treaty and subsequent act to enforce it (the Migratory Bird Treaty Act) protected more than 1,100 migratory bird species by making it illegal to pursue, hunt, take, capture, kill or sell live or dead birds, feathers, eggs and nests, except as permitted through hunting regulations for game birds.
What made this treaty particularly inspiring was that President Woodrow Wilson and King George V made the pledge amid the chaos of World War I. Soon after, the treaty was used as a model for similar agreements with Mexico (in 1936), Japan (1972) and Russia (1976). The Migratory Bird Treaties and subsequent international collaborations to conserve birds show how, when taken together, these global and hemispheric actions are far more than the sum of their parts. Collectively, these efforts are paving a path forward to protect birds and the ecosystems on which both birds and people depend.
International efforts are necessary to conserve migratory birds because birds don’t recognize geopolitical borders. Over the course of a year, songbirds, like the magnolia warbler, may spend 80 days breeding in the boreal forests of the northern U.S. and Canada, 30 days at resting and refueling sites during migration, and over 200 days overwintering in Latin American countries like Mexico, Belize, Guatemala and Honduras.
Looking back at the last 100 years, it’s heartening to see that where we acted together, we had success. The plume trade was virtually halted, and populations of herons, egrets and ibises rebounded. Waterfowl, too, have benefited tremendously from multinational habitat restoration and careful hunting management, in part guided by the North American Waterfowl Management Plan and transformative legislation like the North American Wetlands Conservation Act that leveraged billions in funding for restoration and conservation of over 30 million acres in the U.S., Canada and Mexico. Hunters and sportspeople were instrumental in this success, as they became the key drivers of conservation. Our investments have paid off: Populations of waterfowl and other waterbirds have increased, and the wetlands protected along the way now keep our drinking water clean and reduce flood risk.
But we need to do more. Unlike a century ago, when hunting decimated bird populations, today’s threats are more insidious. Birds are often collateral damage when habitat is lost and exotic species invade, and they are further endangered by collision with buildings and other structures, contamination, and even seemingly innocuous choices like letting our cats roam freely outdoors. These pervasive threats span geopolitical borders and, consequently, are best addressed through coordinated international action. Indeed, the recent “The State of North America’s Birds 2016” report indicates that without conservation action, over one-third of all North American bird species are at risk of extinction.
As we approach the next 100 years of conservation, we must remember that bird conservation is not only about birds; it’s about people, as well. We derive so many benefits from healthy bird populations, including pollination, seed dispersal, insect control and other ecosystem services. Birds also help us to understand the world around us and connect us with nature. And birds are a critical economic resource as well. Activities like hunting and birdwatching contribute billions of dollars to the U.S. economy alone. In fact, the North American Bird Conservation Initiative was originally created by the governments of Canada, the U.S. and Mexico in 1999 to recognize birds as an international “natural economic resource.”
The bottom line is that habitats healthy for birds are also healthy for people. We must work with diverse partners and stakeholders to identify conservation approaches, like sustainable forestry, that meet the needs of local communities and conserve birds and other species. Looking ahead, I believe that an integrated approach, where social and ecological needs are both accommodated, will be the hallmark of the next century of bird conservation.
Rodewald is the Garvin Professor of ornithology and director of conservation science at the Cornell Lab of Ornithology, faculty in the Department of Natural Resources at Cornell University, and faculty fellow at Cornell University’s Atkinson Center for a Sustainable Future. Views expressed in her column are hers alone and do not represent those of these institutions.
The views expressed by contributors are their own and not the views of The Hill.
The Institute for Computational Sustainability was proud to support Ithaca High School’s Code Red club’s 2016 build and competition. The Code Red Robotics Team 639 is part of the FIRST (For Inspiration and Recognition of Science and Technology) groups founded by Dean Kamen. In 2016 more than 75,000 students in more than 3,000 teams in 10 countries participated in 74 regional competitions that led to 1 championship competition. ICS supported the local Ithaca High School team Code Red during the six-week build session when students and their mentors design, build and test a robot, and then participate in successive rounds of competition. Institute for Computational Sustainability especially admires the teamwork and creative problem-solving fostered by Code Red, which will prepare students well for further study in computer science, engineering, and design.
At Code Red students get help and advice from Ithaca High tech teachers but also mentors from the world outside the school. Karina Burbank, Code Red’s public relations officer writes, “Many of these mentors are engineers or other business professionals in the community, who give their time to help students during build season. While Code Red is a student-led team and we always make sure our students are involved and make major decisions, our mentors are vital to our success. They advise and guide students, helping every team member learn as much as possible, each day. This past season, we were lucky to have 16 community mentors.”
Luvelle Brown, superintendent of Ithaca City School District commented on the relevancy of the club,” Code Red Robotics has done the best job of any program in the district that I can think of at truly integrating technology, at relationship-building between students and adults, and relevancy, making the curriculum come alive in a way that engages and makes the work important to young people.”
Anyone is welcome to join, and students can participate in a number of ways, so it’s not just for tech-trained nerds or kids who are super smart in math and physics. Many kids participate all four years of high school even though the time commitment during the six weeks of ‘build’ commits them to at least 31 hours each week over and above their regular classes and homework.
From January to March, Code Red designs and builds their robot to meet the challenge sent out by FIRST, and develops their strategy to use during competitions. Karina explains, “This challenge is different every year, but it’s always some sort of game that has two alliances of three robots competing against each other. This year’s game was called FIRST Stronghold, and our robot had to cross a series of defenses like a drawbridge, portcullis, and other similar things, as well as shoot balls into a goal, and pull itself up on a five-foot high bar.”
For some students, it’s all about winning, and Code Red did very well during 2016. Karina let us know that, ”This year, we won the Fingerlakes Regional competition, as well as the Engineering Inspiration Award at Fingerlakes Regional (which celebrates outreach and education about STEM and robotics in a community), both of which qualified us for World Championships. At World Championships, we placed 6th in our division, and got to our division’s semi-finals as an alliance leader. (There are so many teams at champs that they are divided into several, simultaneously playing divisions.) This puts us in the top 2% of teams world-wide this year.”
The FIRST competitions are designed so teams must form alliances and cooperate with other teams in order to win. A few years ago Code Red was instrumental in helping nearby Trumansburg form a team, and this year Trumansburg’s team brought a great robot to competition. In April Team 639:Code Red Robotics posted on their Facebook page,” We’re going to elimination matches!! We won our last qualification match, placing us in 6th place in our division! This means that tomorrow, we will be able to select our own alliance to compete with in the finals. We think that this is the first time in Code Red history that we’ve been alliance captain in Champs elimination rounds, though we’ve participated as part of an alliance before.’ The Ithaca Journal reported that Trumansburg’s team co-founder Kevin Griswold was grateful to Ithaca’s Code Red team, “In our first year when we first formed the team, the whole group of us attended Code Red’s first meeting just to see how they ran things and to use them in a leadership role and as a role model for how we wanted to project ourselves,” Griswold said. “It’s only grown from there. Last year, we collaborated on a few practice runs to compete and practice against each other and did that again this year at the end of build season just to have another team to practice with.”
Code Red is not just about building robots. Karina helped us understand the wider scope of the club’s activities. “Outside of the build season, we do a lot of community service and demos. This past year, there were 56 students on our team and we estimate that in total, including the individual community service we ask our team members to do, we did about 2,360 hours of community service. We also frequently do demonstrations and presentations about our robot and engineering, especially to young kids at the Sciencenter, 4-H, and local schools. Community service and outreach is a huge part of our team, and we always try to be as active in our community as possible, and help spread knowledge about Science, Technology, Engineering, and Math (STEM) to our community.” Casey Dill, a Code Red and IHS alumnus says, ”It’s a team that’s really trying to help the community, teaching people and getting the younger generation excited about robotics and engineering before they ever get to the high school. Younger kids get a light in their eyes when they see what’s going on.”
Professor Gomes likes to support projects that are both competitive and cooperative, like Code Red. She is recognized as a leader in the computational sustainability field, having received a second 10-million dollar grant from the National Science Foundation to develop the field of Computational Sustainability, so her own projects are both cutting-edge and collaborative, bringing together multiple scientists and institutions from a wide range of fields.
To see Team 639 in action watch this documentary produced by Ithaca College students.