Machine learning

Subterms

    More stories

    • 125 Shares99 Views

      in Energy

      How can we reduce the carbon footprint of global computing?

      by

      The voracious appetite for energy from the world’s computers and communications technology presents a clear threat for the globe’s warming climate. That was the blunt assessment from presenters in the intensive two-day Climate Implications of Computing and Communications workshop held on March 3 and 4, hosted by MIT’s Climate and Sustainability Consortium (MCSC), MIT-IBM Watson AI Lab, and the Schwarzman College of Computing.

      The virtual event featured rich discussions and highlighted opportunities for collaboration among an interdisciplinary group of MIT faculty and researchers and industry leaders across multiple sectors — underscoring the power of academia and industry coming together.

      “If we continue with the existing trajectory of compute energy, by 2040, we are supposed to hit the world’s energy production capacity. The increase in compute energy and demand has been increasing at a much faster rate than the world energy production capacity increase,” said Bilge Yildiz, the Breene M. Kerr Professor in the MIT departments of Nuclear Science and Engineering and Materials Science and Engineering, one of the workshop’s 18 presenters. This computing energy projection draws from the Semiconductor Research Corporations’s decadal report.To cite just one example: Information and communications technology already account for more than 2 percent of global energy demand, which is on a par with the aviation industries emissions from fuel.“We are the very beginning of this data-driven world. We really need to start thinking about this and act now,” said presenter Evgeni Gousev, senior director at Qualcomm.  Innovative energy-efficiency optionsTo that end, the workshop presentations explored a host of energy-efficiency options, including specialized chip design, data center architecture, better algorithms, hardware modifications, and changes in consumer behavior. Industry leaders from AMD, Ericsson, Google, IBM, iRobot, NVIDIA, Qualcomm, Tertill, Texas Instruments, and Verizon outlined their companies’ energy-saving programs, while experts from across MIT provided insight into current research that could yield more efficient computing.Panel topics ranged from “Custom hardware for efficient computing” to “Hardware for new architectures” to “Algorithms for efficient computing,” among others.

      Visual representation of the conversation during the workshop session entitled “Energy Efficient Systems.”

      Image: Haley McDevitt

      Previous item
      Next item

      The goal, said Yildiz, is to improve energy efficiency associated with computing by more than a million-fold.“I think part of the answer of how we make computing much more sustainable has to do with specialized architectures that have very high level of utilization,” said Darío Gil, IBM senior vice president and director of research, who stressed that solutions should be as “elegant” as possible.  For example, Gil illustrated an innovative chip design that uses vertical stacking to reduce the distance data has to travel, and thus reduces energy consumption. Surprisingly, more effective use of tape — a traditional medium for primary data storage — combined with specialized hard drives (HDD), can yield a dramatic savings in carbon dioxide emissions.Gil and presenters Bill Dally, chief scientist and senior vice president of research of NVIDIA; Ahmad Bahai, CTO of Texas Instruments; and others zeroed in on storage. Gil compared data to a floating iceberg in which we can have fast access to the “hot data” of the smaller visible part while the “cold data,” the large underwater mass, represents data that tolerates higher latency. Think about digital photo storage, Gil said. “Honestly, are you really retrieving all of those photographs on a continuous basis?” Storage systems should provide an optimized mix of of HDD for hot data and tape for cold data based on data access patterns.Bahai stressed the significant energy saving gained from segmenting standby and full processing. “We need to learn how to do nothing better,” he said. Dally spoke of mimicking the way our brain wakes up from a deep sleep, “We can wake [computers] up much faster, so we don’t need to keep them running in full speed.”Several workshop presenters spoke of a focus on “sparsity,” a matrix in which most of the elements are zero, as a way to improve efficiency in neural networks. Or as Dally said, “Never put off till tomorrow, where you could put off forever,” explaining efficiency is not “getting the most information with the fewest bits. It’s doing the most with the least energy.”Holistic and multidisciplinary approaches“We need both efficient algorithms and efficient hardware, and sometimes we need to co-design both the algorithm and the hardware for efficient computing,” said Song Han, a panel moderator and assistant professor in the Department of Electrical Engineering and Computer Science (EECS) at MIT.Some presenters were optimistic about innovations already underway. According to Ericsson’s research, as much as 15 percent of the carbon emissions globally can be reduced through the use of existing solutions, noted Mats Pellbäck Scharp, head of sustainability at Ericsson. For example, GPUs are more efficient than CPUs for AI, and the progression from 3G to 5G networks boosts energy savings.“5G is the most energy efficient standard ever,” said Scharp. “We can build 5G without increasing energy consumption.”Companies such as Google are optimizing energy use at their data centers through improved design, technology, and renewable energy. “Five of our data centers around the globe are operating near or above 90 percent carbon-free energy,” said Jeff Dean, Google’s senior fellow and senior vice president of Google Research.Yet, pointing to the possible slowdown in the doubling of transistors in an integrated circuit — or Moore’s Law — “We need new approaches to meet this compute demand,” said Sam Naffziger, AMD senior vice president, corporate fellow, and product technology architect. Naffziger spoke of addressing performance “overkill.” For example, “we’re finding in the gaming and machine learning space we can make use of lower-precision math to deliver an image that looks just as good with 16-bit computations as with 32-bit computations, and instead of legacy 32b math to train AI networks, we can use lower-energy 8b or 16b computations.”

      Visual representation of the conversation during the workshop session entitled “Wireless, networked, and distributed systems.”

      Image: Haley McDevitt

      Previous item
      Next item

      Other presenters singled out compute at the edge as a prime energy hog.“We also have to change the devices that are put in our customers’ hands,” said Heidi Hemmer, senior vice president of engineering at Verizon. As we think about how we use energy, it is common to jump to data centers — but it really starts at the device itself, and the energy that the devices use. Then, we can think about home web routers, distributed networks, the data centers, and the hubs. “The devices are actually the least energy-efficient out of that,” concluded Hemmer.Some presenters had different perspectives. Several called for developing dedicated silicon chipsets for efficiency. However, panel moderator Muriel Medard, the Cecil H. Green Professor in EECS, described research at MIT, Boston University, and Maynooth University on the GRAND (Guessing Random Additive Noise Decoding) chip, saying, “rather than having obsolescence of chips as the new codes come in and in different standards, you can use one chip for all codes.”Whatever the chip or new algorithm, Helen Greiner, CEO of Tertill (a weeding robot) and co-founder of iRobot, emphasized that to get products to market, “We have to learn to go away from wanting to get the absolute latest and greatest, the most advanced processor that usually is more expensive.” She added, “I like to say robot demos are a dime a dozen, but robot products are very infrequent.”Greiner emphasized consumers can play a role in pushing for more energy-efficient products — just as drivers began to demand electric cars.Dean also sees an environmental role for the end user.“We have enabled our cloud customers to select which cloud region they want to run their computation in, and they can decide how important it is that they have a low carbon footprint,” he said, also citing other interfaces that might allow consumers to decide which air flights are more efficient or what impact installing a solar panel on their home would have.However, Scharp said, “Prolonging the life of your smartphone or tablet is really the best climate action you can do if you want to reduce your digital carbon footprint.”Facing increasing demandsDespite their optimism, the presenters acknowledged the world faces increasing compute demand from machine learning, AI, gaming, and especially, blockchain. Panel moderator Vivienne Sze, associate professor in EECS, noted the conundrum.“We can do a great job in making computing and communication really efficient. But there is this tendency that once things are very efficient, people use more of it, and this might result in an overall increase in the usage of these technologies, which will then increase our overall carbon footprint,” Sze said.Presenters saw great potential in academic/industry partnerships, particularly from research efforts on the academic side. “By combining these two forces together, you can really amplify the impact,” concluded Gousev.Presenters at the Climate Implications of Computing and Communications workshop also included: Joel Emer, professor of the practice in EECS at MIT; David Perreault, the Joseph F. and Nancy P. Keithley Professor of EECS at MIT; Jesús del Alamo, MIT Donner Professor and professor of electrical engineering in EECS at MIT; Heike Riel, IBM Fellow and head science and technology at IBM; and Takashi Ando, principal research staff member at IBM Research. The recorded workshop sessions are available on YouTube. More

    • 125 Shares159 Views

      in Energy

      Machine learning, harnessed to extreme computing, aids fusion energy development

      by

      MIT research scientists Pablo Rodriguez-Fernandez and Nathan Howard have just completed one of the most demanding calculations in fusion science — predicting the temperature and density profiles of a magnetically confined plasma via first-principles simulation of plasma turbulence. Solving this problem by brute force is beyond the capabilities of even the most advanced supercomputers. Instead, the researchers used an optimization methodology developed for machine learning to dramatically reduce the CPU time required while maintaining the accuracy of the solution.

      Fusion energyFusion offers the promise of unlimited, carbon-free energy through the same physical process that powers the sun and the stars. It requires heating the fuel to temperatures above 100 million degrees, well above the point where the electrons are stripped from their atoms, creating a form of matter called plasma. On Earth, researchers use strong magnetic fields to isolate and insulate the hot plasma from ordinary matter. The stronger the magnetic field, the better the quality of the insulation that it provides.

      Rodriguez-Fernandez and Howard have focused on predicting the performance expected in the SPARC device, a compact, high-magnetic-field fusion experiment, currently under construction by the MIT spin-out company Commonwealth Fusion Systems (CFS) and researchers from MIT’s Plasma Science and Fusion Center. While the calculation required an extraordinary amount of computer time, over 8 million CPU-hours, what was remarkable was not how much time was used, but how little, given the daunting computational challenge.

      The computational challenge of fusion energyTurbulence, which is the mechanism for most of the heat loss in a confined plasma, is one of the science’s grand challenges and the greatest problem remaining in classical physics. The equations that govern fusion plasmas are well known, but analytic solutions are not possible in the regimes of interest, where nonlinearities are important and solutions encompass an enormous range of spatial and temporal scales. Scientists resort to solving the equations by numerical simulation on computers. It is no accident that fusion researchers have been pioneers in computational physics for the last 50 years.

      One of the fundamental problems for researchers is reliably predicting plasma temperature and density given only the magnetic field configuration and the externally applied input power. In confinement devices like SPARC, the external power and the heat input from the fusion process are lost through turbulence in the plasma. The turbulence itself is driven by the difference in the extremely high temperature of the plasma core and the relatively cool temperatures of the plasma edge (merely a few million degrees). Predicting the performance of a self-heated fusion plasma therefore requires a calculation of the power balance between the fusion power input and the losses due to turbulence.

      These calculations generally start by assuming plasma temperature and density profiles at a particular location, then computing the heat transported locally by turbulence. However, a useful prediction requires a self-consistent calculation of the profiles across the entire plasma, which includes both the heat input and turbulent losses. Directly solving this problem is beyond the capabilities of any existing computer, so researchers have developed an approach that stitches the profiles together from a series of demanding but tractable local calculations. This method works, but since the heat and particle fluxes depend on multiple parameters, the calculations can be very slow to converge.

      However, techniques emerging from the field of machine learning are well suited to optimize just such a calculation. Starting with a set of computationally intensive local calculations run with the full-physics, first-principles CGYRO code (provided by a team from General Atomics led by Jeff Candy) Rodriguez-Fernandez and Howard fit a surrogate mathematical model, which was used to explore and optimize a search within the parameter space. The results of the optimization were compared to the exact calculations at each optimum point, and the system was iterated to a desired level of accuracy. The researchers estimate that the technique reduced the number of runs of the CGYRO code by a factor of four.

      New approach increases confidence in predictionsThis work, described in a recent publication in the journal Nuclear Fusion, is the highest fidelity calculation ever made of the core of a fusion plasma. It refines and confirms predictions made with less demanding models. Professor Jonathan Citrin, of the Eindhoven University of Technology and leader of the fusion modeling group for DIFFER, the Dutch Institute for Fundamental Energy Research, commented: “The work significantly accelerates our capabilities in more routinely performing ultra-high-fidelity tokamak scenario prediction. This algorithm can help provide the ultimate validation test of machine design or scenario optimization carried out with faster, more reduced modeling, greatly increasing our confidence in the outcomes.” 

      In addition to increasing confidence in the fusion performance of the SPARC experiment, this technique provides a roadmap to check and calibrate reduced physics models, which run with a small fraction of the computational power. Such models, cross-checked against the results generated from turbulence simulations, will provide a reliable prediction before each SPARC discharge, helping to guide experimental campaigns and improving the scientific exploitation of the device. It can also be used to tweak and improve even simple data-driven models, which run extremely quickly, allowing researchers to sift through enormous parameter ranges to narrow down possible experiments or possible future machines.

      The research was funded by CFS, with computational support from the National Energy Research Scientific Computing Center, a U.S. Department of Energy Office of Science User Facility. More

    • 188 Shares199 Views

      in Energy

      What choices does the world need to make to keep global warming below 2 C?

      by

      When the 2015 Paris Agreement set a long-term goal of keeping global warming “well below 2 degrees Celsius, compared to pre-industrial levels” to avoid the worst impacts of climate change, it did not specify how its nearly 200 signatory nations could collectively achieve that goal. Each nation was left to its own devices to reduce greenhouse gas emissions in alignment with the 2 C target. Now a new modeling strategy developed at the MIT Joint Program on the Science and Policy of Global Change that explores hundreds of potential future development pathways provides new insights on the energy and technology choices needed for the world to meet that target.

      Described in a study appearing in the journal Earth’s Future, the new strategy combines two well-known computer modeling techniques to scope out the energy and technology choices needed over the coming decades to reduce emissions sufficiently to achieve the Paris goal.

      The first technique, Monte Carlo analysis, quantifies uncertainty levels for dozens of energy and economic indicators including fossil fuel availability, advanced energy technology costs, and population and economic growth; feeds that information into a multi-region, multi-economic-sector model of the world economy that captures the cross-sectoral impacts of energy transitions; and runs that model hundreds of times to estimate the likelihood of different outcomes. The MIT study focuses on projections through the year 2100 of economic growth and emissions for different sectors of the global economy, as well as energy and technology use.

      The second technique, scenario discovery, uses machine learning tools to screen databases of model simulations in order to identify outcomes of interest and their conditions for occurring. The MIT study applies these tools in a unique way by combining them with the Monte Carlo analysis to explore how different outcomes are related to one another (e.g., do low-emission outcomes necessarily involve large shares of renewable electricity?). This approach can also identify individual scenarios, out of the hundreds explored, that result in specific combinations of outcomes of interest (e.g., scenarios with low emissions, high GDP growth, and limited impact on electricity prices), and also provide insight into the conditions needed for that combination of outcomes.

      Using this unique approach, the MIT Joint Program researchers find several possible patterns of energy and technology development under a specified long-term climate target or economic outcome.

      “This approach shows that there are many pathways to a successful energy transition that can be a win-win for the environment and economy,” says Jennifer Morris, an MIT Joint Program research scientist and the study’s lead author. “Toward that end, it can be used to guide decision-makers in government and industry to make sound energy and technology choices and avoid biases in perceptions of what ’needs’ to happen to achieve certain outcomes.”

      For example, while achieving the 2 C goal, the global level of combined wind and solar electricity generation by 2050 could be less than three times or more than 12 times the current level (which is just over 2,000 terawatt hours). These are very different energy pathways, but both can be consistent with the 2 C goal. Similarly, there are many different energy mixes that can be consistent with maintaining high GDP growth in the United States while also achieving the 2 C goal, with different possible roles for renewables, natural gas, carbon capture and storage, and bioenergy. The study finds renewables to be the most robust electricity investment option, with sizable growth projected under each of the long-term temperature targets explored.

      The researchers also find that long-term climate targets have little impact on economic output for most economic sectors through 2050, but do require each sector to significantly accelerate reduction of its greenhouse gas emissions intensity (emissions per unit of economic output) so as to reach near-zero levels by midcentury.

      “Given the range of development pathways that can be consistent with meeting a 2 degrees C goal, policies that target only specific sectors or technologies can unnecessarily narrow the solution space, leading to higher costs,” says former MIT Joint Program Co-Director John Reilly, a co-author of the study. “Our findings suggest that policies designed to encourage a portfolio of technologies and sectoral actions can be a wise strategy that hedges against risks.”

      The research was supported by the U.S. Department of Energy Office of Science. More

    • 138 Shares99 Views

      in Energy

      Engineers enlist AI to help scale up advanced solar cell manufacturing

      by

      Perovskites are a family of materials that are currently the leading contender to potentially replace today’s silicon-based solar photovoltaics. They hold the promise of panels that are far thinner and lighter, that could be made with ultra-high throughput at room temperature instead of at hundreds of degrees, and that are cheaper and easier to transport and install. But bringing these materials from controlled laboratory experiments into a product that can be manufactured competitively has been a long struggle.

      Manufacturing perovskite-based solar cells involves optimizing at least a dozen or so variables at once, even within one particular manufacturing approach among many possibilities. But a new system based on a novel approach to machine learning could speed up the development of optimized production methods and help make the next generation of solar power a reality.

      The system, developed by researchers at MIT and Stanford University over the last few years, makes it possible to integrate data from prior experiments, and information based on personal observations by experienced workers, into the machine learning process. This makes the outcomes more accurate and has already led to the manufacturing of perovskite cells with an energy conversion efficiency of 18.5 percent, a competitive level for today’s market.

      The research is reported today in the journal Joule, in a paper by MIT professor of mechanical engineering Tonio Buonassisi, Stanford professor of materials science and engineering Reinhold Dauskardt, recent MIT research assistant Zhe Liu, Stanford doctoral graduate Nicholas Rolston, and three others.

      Perovskites are a group of layered crystalline compounds defined by the configuration of the atoms in their crystal lattice. There are thousands of such possible compounds and many different ways of making them. While most lab-scale development of perovskite materials uses a spin-coating technique, that’s not practical for larger-scale manufacturing, so companies and labs around the world have been searching for ways of translating these lab materials into a practical, manufacturable product.

      “There’s always a big challenge when you’re trying to take a lab-scale process and then transfer it to something like a startup or a manufacturing line,” says Rolston, who is now an assistant professor at Arizona State University. The team looked at a process that they felt had the greatest potential, a method called rapid spray plasma processing, or RSPP.

      The manufacturing process would involve a moving roll-to-roll surface, or series of sheets, on which the precursor solutions for the perovskite compound would be sprayed or ink-jetted as the sheet rolled by. The material would then move on to a curing stage, providing a rapid and continuous output “with throughputs that are higher than for any other photovoltaic technology,” Rolston says.

      “The real breakthrough with this platform is that it would allow us to scale in a way that no other material has allowed us to do,” he adds. “Even materials like silicon require a much longer timeframe because of the processing that’s done. Whereas you can think of [this approach as more] like spray painting.”

      Within that process, at least a dozen variables may affect the outcome, some of them more controllable than others. These include the composition of the starting materials, the temperature, the humidity, the speed of the processing path, the distance of the nozzle used to spray the material onto a substrate, and the methods of curing the material. Many of these factors can interact with each other, and if the process is in open air, then humidity, for example, may be uncontrolled. Evaluating all possible combinations of these variables through experimentation is impossible, so machine learning was needed to help guide the experimental process.

      But while most machine-learning systems use raw data such as measurements of the electrical and other properties of test samples, they don’t typically incorporate human experience such as qualitative observations made by the experimenters of the visual and other properties of the test samples, or information from other experiments reported by other researchers. So, the team found a way to incorporate such outside information into the machine learning model, using a probability factor based on a mathematical technique called Bayesian Optimization.

      Using the system, he says, “having a model that comes from experimental data, we can find out trends that we weren’t able to see before.” For example, they initially had trouble adjusting for uncontrolled variations in humidity in their ambient setting. But the model showed them “that we could overcome our humidity challenges by changing the temperature, for instance, and by changing some of the other knobs.”

      The system now allows experimenters to much more rapidly guide their process in order to optimize it for a given set of conditions or required outcomes. In their experiments, the team focused on optimizing the power output, but the system could also be used to simultaneously incorporate other criteria, such as cost and durability — something members of the team are continuing to work on, Buonassisi says.

      The researchers were encouraged by the Department of Energy, which sponsored the work, to commercialize the technology, and they’re currently focusing on tech transfer to existing perovskite manufacturers. “We are reaching out to companies now,” Buonassisi says, and the code they developed has been made freely available through an open-source server. “It’s now on GitHub, anyone can download it, anyone can run it,” he says. “We’re happy to help companies get started in using our code.”

      Already, several companies are gearing up to produce perovskite-based solar panels, even though they are still working out the details of how to produce them, says Liu, who is now at the Northwestern Polytechnical University in Xi’an, China. He says companies there are not yet doing large-scale manufacturing, but instead starting with smaller, high-value applications such as building-integrated solar tiles where appearance is important. Three of these companies “are on track or are being pushed by investors to manufacture 1 meter by 2-meter rectangular modules [comparable to today’s most common solar panels], within two years,” he says.

      ‘The problem is, they don’t have a consensus on what manufacturing technology to use,” Liu says. The RSPP method, developed at Stanford, “still has a good chance” to be competitive, he says. And the machine learning system the team developed could prove to be important in guiding the optimization of whatever process ends up being used.

      “The primary goal was to accelerate the process, so it required less time, less experiments, and less human hours to develop something that is usable right away, for free, for industry,” he says.

      “Existing work on machine-learning-driven perovskite PV fabrication largely focuses on spin-coating, a lab-scale technique,” says Ted Sargent, University Professor at the University of Toronto, who was not associated with this work, which he says demonstrates “a workflow that is readily adapted to the deposition techniques that dominate the thin-film industry. Only a handful of groups have the simultaneous expertise in engineering and computation to drive such advances.” Sargent adds that this approach “could be an exciting advance for the manufacture of a broader family of materials” including LEDs, other PV technologies, and graphene, “in short, any industry that uses some form of vapor or vacuum deposition.” 

      The team also included Austin Flick and Thomas Colburn at Stanford and Zekun Ren at the Singapore-MIT Alliance for Science and Technology (SMART). In addition to the Department of Energy, the work was supported by a fellowship from the MIT Energy Initiative, the Graduate Research Fellowship Program from the National Science Foundation, and the SMART program. More

    • 50 Shares159 Views

      in Environment

      Computing our climate future

      by

      On Monday, MIT announced five multiyear flagship projects in the first-ever Climate Grand Challenges, a new initiative to tackle complex climate problems and deliver breakthrough solutions to the world as quickly as possible. This article is the first in a five-part series highlighting the most promising concepts to emerge from the competition, and the interdisciplinary research teams behind them.

      With improvements to computer processing power and an increased understanding of the physical equations governing the Earth’s climate, scientists are continually working to refine climate models and improve their predictive power. But the tools they’re refining were originally conceived decades ago with only scientists in mind. When it comes to developing tangible climate action plans, these models remain inscrutable to the policymakers, public safety officials, civil engineers, and community organizers who need their predictive insight most.

      “What you end up having is a gap between what’s typically used in practice, and the real cutting-edge science,” says Noelle Selin, a professor in the Institute for Data, Systems and Society and the Department of Earth, Atmospheric and Planetary Sciences (EAPS), and co-lead with Professor Raffaele Ferrari on the MIT Climate Grand Challenges flagship project “Bringing Computation to the Climate Crisis.” “How can we use new computational techniques, new understandings, new ways of thinking about modeling, to really bridge that gap between state-of-the-art scientific advances and modeling, and people who are actually needing to use these models?”

      Using this as a driving question, the team won’t just be trying to refine current climate models, they’re building a new one from the ground up.

      This kind of game-changing advancement is exactly what the MIT Climate Grand Challenges is looking for, which is why the proposal has been named one of the five flagship projects in the ambitious Institute-wide program aimed at tackling the climate crisis. The proposal, which was selected from 100 submissions and was among 27 finalists, will receive additional funding and support to further their goal of reimagining the climate modeling system. It also brings together contributors from across the Institute, including the MIT Schwarzman College of Computing, the School of Engineering, and the Sloan School of Management.

      When it comes to pursuing high-impact climate solutions that communities around the world can use, “it’s great to do it at MIT,” says Ferrari, EAPS Cecil and Ida Green Professor of Oceanography. “You’re not going to find many places in the world where you have the cutting-edge climate science, the cutting-edge computer science, and the cutting-edge policy science experts that we need to work together.”

      The climate model of the future

      The proposal builds on work that Ferrari began three years ago as part of a joint project with Caltech, the Naval Postgraduate School, and NASA’s Jet Propulsion Lab. Called the Climate Modeling Alliance (CliMA), the consortium of scientists, engineers, and applied mathematicians is constructing a climate model capable of more accurately projecting future changes in critical variables, such as clouds in the atmosphere and turbulence in the ocean, with uncertainties at least half the size of those in existing models.

      To do this, however, requires a new approach. For one thing, current models are too coarse in resolution — at the 100-to-200-kilometer scale — to resolve small-scale processes like cloud cover, rainfall, and sea ice extent. But also, explains Ferrari, part of this limitation in resolution is due to the fundamental architecture of the models themselves. The languages most global climate models are coded in were first created back in the 1960s and ’70s, largely by scientists for scientists. Since then, advances in computing driven by the corporate world and computer gaming have given rise to dynamic new computer languages, powerful graphics processing units, and machine learning.

      For climate models to take full advantage of these advancements, there’s only one option: starting over with a modern, more flexible language. Written in Julia, a part of Julialab’s Scientific Machine Learning technology, and spearheaded by Alan Edelman, a professor of applied mathematics in MIT’s Department of Mathematics, CliMA will be able to harness far more data than the current models can handle.

      “It’s been real fun finally working with people in computer science here at MIT,” Ferrari says. “Before it was impossible, because traditional climate models are in a language their students can’t even read.”

      The result is what’s being called the “Earth digital twin,” a climate model that can simulate global conditions on a large scale. This on its own is an impressive feat, but the team wants to take this a step further with their proposal.

      “We want to take this large-scale model and create what we call an ‘emulator’ that is only predicting a set of variables of interest, but it’s been trained on the large-scale model,” Ferrari explains. Emulators are not new technology, but what is new is that these emulators, being referred to as the “Earth digital cousins,” will take advantage of machine learning.

      “Now we know how to train a model if we have enough data to train them on,” says Ferrari. Machine learning for projects like this has only become possible in recent years as more observational data become available, along with improved computer processing power. The goal is to create smaller, more localized models by training them using the Earth digital twin. Doing so will save time and money, which is key if the digital cousins are going to be usable for stakeholders, like local governments and private-sector developers.

      Adaptable predictions for average stakeholders

      When it comes to setting climate-informed policy, stakeholders need to understand the probability of an outcome within their own regions — in the same way that you would prepare for a hike differently if there’s a 10 percent chance of rain versus a 90 percent chance. The smaller Earth digital cousin models will be able to do things the larger model can’t do, like simulate local regions in real time and provide a wider range of probabilistic scenarios.

      “Right now, if you wanted to use output from a global climate model, you usually would have to use output that’s designed for general use,” says Selin, who is also the director of the MIT Technology and Policy Program. With the project, the team can take end-user needs into account from the very beginning while also incorporating their feedback and suggestions into the models, helping to “democratize the idea of running these climate models,” as she puts it. Doing so means building an interactive interface that eventually will give users the ability to change input values and run the new simulations in real time. The team hopes that, eventually, the Earth digital cousins could run on something as ubiquitous as a smartphone, although developments like that are currently beyond the scope of the project.

      The next thing the team will work on is building connections with stakeholders. Through participation of other MIT groups, such as the Joint Program on the Science and Policy of Global Change and the Climate and Sustainability Consortium, they hope to work closely with policymakers, public safety officials, and urban planners to give them predictive tools tailored to their needs that can provide actionable outputs important for planning. Faced with rising sea levels, for example, coastal cities could better visualize the threat and make informed decisions about infrastructure development and disaster preparedness; communities in drought-prone regions could develop long-term civil planning with an emphasis on water conservation and wildfire resistance.

      “We want to make the modeling and analysis process faster so people can get more direct and useful feedback for near-term decisions,” she says.

      The final piece of the challenge is to incentivize students now so that they can join the project and make a difference. Ferrari has already had luck garnering student interest after co-teaching a class with Edelman and seeing the enthusiasm students have about computer science and climate solutions.

      “We’re intending in this project to build a climate model of the future,” says Selin. “So it seems really appropriate that we would also train the builders of that climate model.” More

    • 125 Shares179 Views

      in Environment

      Improving predictions of sea level rise for the next century

      by

      When we think of climate change, one of the most dramatic images that comes to mind is the loss of glacial ice. As the Earth warms, these enormous rivers of ice become a casualty of the rising temperatures. But, as ice sheets retreat, they also become an important contributor to one the more dangerous outcomes of climate change: sea-level rise. At MIT, an interdisciplinary team of scientists is determined to improve sea level rise predictions for the next century, in part by taking a closer look at the physics of ice sheets.

      Last month, two research proposals on the topic, led by Brent Minchew, the Cecil and Ida Green Career Development Professor in the Department of Earth, Atmospheric and Planetary Sciences (EAPS), were announced as finalists in the MIT Climate Grand Challenges initiative. Launched in July 2020, Climate Grand Challenges fielded almost 100 project proposals from collaborators across the Institute who heeded the bold charge: to develop research and innovations that will deliver game-changing advances in the world’s efforts to address the climate challenge.

      As finalists, Minchew and his collaborators from the departments of Urban Studies and Planning, Economics, Civil and Environmental Engineering, the Haystack Observatory, and external partners, received $100,000 to develop their research plans. A subset of the 27 proposals tapped as finalists will be announced next month, making up a portfolio of multiyear “flagship” projects receiving additional funding and support.

      One goal of both Minchew proposals is to more fully understand the most fundamental processes that govern rapid changes in glacial ice, and to use that understanding to build next-generation models that are more predictive of ice sheet behavior as they respond to, and influence, climate change.

      “We need to develop more accurate and computationally efficient models that provide testable projections of sea-level rise over the coming decades. To do so quickly, we want to make better and more frequent observations and learn the physics of ice sheets from these data,” says Minchew. “For example, how much stress do you have to apply to ice before it breaks?”

      Currently, Minchew’s Glacier Dynamics and Remote Sensing group uses satellites to observe the ice sheets on Greenland and Antarctica primarily with interferometric synthetic aperture radar (InSAR). But the data are often collected over long intervals of time, which only gives them “before and after” snapshots of big events. By taking more frequent measurements on shorter time scales, such as hours or days, they can get a more detailed picture of what is happening in the ice.

      “Many of the key unknowns in our projections of what ice sheets are going to look like in the future, and how they’re going to evolve, involve the dynamics of glaciers, or our understanding of how the flow speed and the resistances to flow are related,” says Minchew.

      At the heart of the two proposals is the creation of SACOS, the Stratospheric Airborne Climate Observatory System. The group envisions developing solar-powered drones that can fly in the stratosphere for months at a time, taking more frequent measurements using a new lightweight, low-power radar and other high-resolution instrumentation. They also propose air-dropping sensors directly onto the ice, equipped with seismometers and GPS trackers to measure high-frequency vibrations in the ice and pinpoint the motions of its flow.

      How glaciers contribute to sea level rise

      Current climate models predict an increase in sea levels over the next century, but by just how much is still unclear. Estimates are anywhere from 20 centimeters to two meters, which is a large difference when it comes to enacting policy or mitigation. Minchew points out that response measures will be different, depending on which end of the scale it falls toward. If it’s closer to 20 centimeters, coastal barriers can be built to protect low-level areas. But with higher surges, such measures become too expensive and inefficient to be viable, as entire portions of cities and millions of people would have to be relocated.

      “If we’re looking at a future where we could get more than a meter of sea level rise by the end of the century, then we need to know about that sooner rather than later so that we can start to plan and to do our best to prepare for that scenario,” he says.

      There are two ways glaciers and ice sheets contribute to rising sea levels: direct melting of the ice and accelerated transport of ice to the oceans. In Antarctica, warming waters melt the margins of the ice sheets, which tends to reduce the resistive stresses and allow ice to flow more quickly to the ocean. This thinning can also cause the ice shelves to be more prone to fracture, facilitating the calving of icebergs — events which sometimes cause even further acceleration of ice flow.

      Using data collected by SACOS, Minchew and his group can better understand what material properties in the ice allow for fracturing and calving of icebergs, and build a more complete picture of how ice sheets respond to climate forces. 

      “What I want is to reduce and quantify the uncertainties in projections of sea level rise out to the year 2100,” he says.

      From that more complete picture, the team — which also includes economists, engineers, and urban planning specialists — can work on developing predictive models and methods to help communities and governments estimate the costs associated with sea level rise, develop sound infrastructure strategies, and spur engineering innovation.

      Understanding glacier dynamics

      More frequent radar measurements and the collection of higher-resolution seismic and GPS data will allow Minchew and the team to develop a better understanding of the broad category of glacier dynamics — including calving, an important process in setting the rate of sea level rise which is currently not well understood.  

      “Some of what we’re doing is quite similar to what seismologists do,” he says. “They measure seismic waves following an earthquake, or a volcanic eruption, or things of this nature and use those observations to better understand the mechanisms that govern these phenomena.”

      Air-droppable sensors will help them collect information about ice sheet movement, but this method comes with drawbacks — like installation and maintenance, which is difficult to do out on a massive ice sheet that is moving and melting. Also, the instruments can each only take measurements at a single location. Minchew equates it to a bobber in water: All it can tell you is how the bobber moves as the waves disturb it.

      But by also taking continuous radar measurements from the air, Minchew’s team can collect observations both in space and in time. Instead of just watching the bobber in the water, they can effectively make a movie of the waves propagating out, as well as visualize processes like iceberg calving happening in multiple dimensions.

      Once the bobbers are in place and the movies recorded, the next step is developing machine learning algorithms to help analyze all the new data being collected. While this data-driven kind of discovery has been a hot topic in other fields, this is the first time it has been applied to glacier research.

      “We’ve developed this new methodology to ingest this huge amount of data,” he says, “and from that create an entirely new way of analyzing the system to answer these fundamental and critically important questions.”  More

    • 125 Shares199 Views

      in Energy

      Using artificial intelligence to find anomalies hiding in massive datasets

      by

      Identifying a malfunction in the nation’s power grid can be like trying to find a needle in an enormous haystack. Hundreds of thousands of interrelated sensors spread across the U.S. capture data on electric current, voltage, and other critical information in real time, often taking multiple recordings per second.

      Researchers at the MIT-IBM Watson AI Lab have devised a computationally efficient method that can automatically pinpoint anomalies in those data streams in real time. They demonstrated that their artificial intelligence method, which learns to model the interconnectedness of the power grid, is much better at detecting these glitches than some other popular techniques.

      Because the machine-learning model they developed does not require annotated data on power grid anomalies for training, it would be easier to apply in real-world situations where high-quality, labeled datasets are often hard to come by. The model is also flexible and can be applied to other situations where a vast number of interconnected sensors collect and report data, like traffic monitoring systems. It could, for example, identify traffic bottlenecks or reveal how traffic jams cascade.

      “In the case of a power grid, people have tried to capture the data using statistics and then define detection rules with domain knowledge to say that, for example, if the voltage surges by a certain percentage, then the grid operator should be alerted. Such rule-based systems, even empowered by statistical data analysis, require a lot of labor and expertise. We show that we can automate this process and also learn patterns from the data using advanced machine-learning techniques,” says senior author Jie Chen, a research staff member and manager of the MIT-IBM Watson AI Lab.

      The co-author is Enyan Dai, an MIT-IBM Watson AI Lab intern and graduate student at the Pennsylvania State University. This research will be presented at the International Conference on Learning Representations.

      Probing probabilities

      The researchers began by defining an anomaly as an event that has a low probability of occurring, like a sudden spike in voltage. They treat the power grid data as a probability distribution, so if they can estimate the probability densities, they can identify the low-density values in the dataset. Those data points which are least likely to occur correspond to anomalies.

      Estimating those probabilities is no easy task, especially since each sample captures multiple time series, and each time series is a set of multidimensional data points recorded over time. Plus, the sensors that capture all that data are conditional on one another, meaning they are connected in a certain configuration and one sensor can sometimes impact others.

      To learn the complex conditional probability distribution of the data, the researchers used a special type of deep-learning model called a normalizing flow, which is particularly effective at estimating the probability density of a sample.

      They augmented that normalizing flow model using a type of graph, known as a Bayesian network, which can learn the complex, causal relationship structure between different sensors. This graph structure enables the researchers to see patterns in the data and estimate anomalies more accurately, Chen explains.

      “The sensors are interacting with each other, and they have causal relationships and depend on each other. So, we have to be able to inject this dependency information into the way that we compute the probabilities,” he says.

      This Bayesian network factorizes, or breaks down, the joint probability of the multiple time series data into less complex, conditional probabilities that are much easier to parameterize, learn, and evaluate. This allows the researchers to estimate the likelihood of observing certain sensor readings, and to identify those readings that have a low probability of occurring, meaning they are anomalies.

      Their method is especially powerful because this complex graph structure does not need to be defined in advance — the model can learn the graph on its own, in an unsupervised manner.

      A powerful technique

      They tested this framework by seeing how well it could identify anomalies in power grid data, traffic data, and water system data. The datasets they used for testing contained anomalies that had been identified by humans, so the researchers were able to compare the anomalies their model identified with real glitches in each system.

      Their model outperformed all the baselines by detecting a higher percentage of true anomalies in each dataset.

      “For the baselines, a lot of them don’t incorporate graph structure. That perfectly corroborates our hypothesis. Figuring out the dependency relationships between the different nodes in the graph is definitely helping us,” Chen says.

      Their methodology is also flexible. Armed with a large, unlabeled dataset, they can tune the model to make effective anomaly predictions in other situations, like traffic patterns.

      Once the model is deployed, it would continue to learn from a steady stream of new sensor data, adapting to possible drift of the data distribution and maintaining accuracy over time, says Chen.

      Though this particular project is close to its end, he looks forward to applying the lessons he learned to other areas of deep-learning research, particularly on graphs.

      Chen and his colleagues could use this approach to develop models that map other complex, conditional relationships. They also want to explore how they can efficiently learn these models when the graphs become enormous, perhaps with millions or billions of interconnected nodes. And rather than finding anomalies, they could also use this approach to improve the accuracy of forecasts based on datasets or streamline other classification techniques.

      This work was funded by the MIT-IBM Watson AI Lab and the U.S. Department of Energy. More

    • 50 Shares159 Views

      in Energy

      More sensitive X-ray imaging

      by

      Scintillators are materials that emit light when bombarded with high-energy particles or X-rays. In medical or dental X-ray systems, they convert incoming X-ray radiation into visible light that can then be captured using film or photosensors. They’re also used for night-vision systems and for research, such as in particle detectors or electron microscopes.

      Researchers at MIT have now shown how one could improve the efficiency of scintillators by at least tenfold, and perhaps even a hundredfold, by changing the material’s surface to create certain nanoscale configurations, such as arrays of wave-like ridges. While past attempts to develop more efficient scintillators have focused on finding new materials, the new approach could in principle work with any of the existing materials.

      Though it will require more time and effort to integrate their scintillators into existing X-ray machines, the team believes that this method might lead to improvements in medical diagnostic X-rays or CT scans, to reduce dose exposure and improve image quality. In other applications, such as X-ray inspection of manufactured parts for quality control, the new scintillators could enable inspections with higher accuracy or at faster speeds.

      The findings are described today in the journal Science, in a paper by MIT doctoral students Charles Roques-Carmes and Nicholas Rivera; MIT professors Marin Soljacic, Steven Johnson, and John Joannopoulos; and 10 others.

      While scintillators have been in use for some 70 years, much of the research in the field has focused on developing new materials that produce brighter or faster light emissions. The new approach instead applies advances in nanotechnology to existing materials. By creating patterns in scintillator materials at a length scale comparable to the wavelengths of the light being emitted, the team found that it was possible to dramatically change the material’s optical properties.

      To make what they coined “nanophotonic scintillators,” Roques-Carmes says, “you can directly make patterns inside the scintillators, or you can glue on another material that would have holes on the nanoscale. The specifics depend on the exact structure and material.” For this research, the team took a scintillator and made holes spaced apart by roughly one optical wavelength, or about 500 nanometers (billionths of a meter).

      “The key to what we’re doing is a general theory and framework we have developed,” Rivera says. This allows the researchers to calculate the scintillation levels that would be produced by any arbitrary configuration of nanophotonic structures. The scintillation process itself involves a series of steps, making it complicated to unravel. The framework the team developed involves integrating three different types of physics, Roques-Carmes says. Using this system they have found a good match between their predictions and the results of their subsequent experiments.

      The experiments showed a tenfold improvement in emission from the treated scintillator. “So, this is something that might translate into applications for medical imaging, which are optical photon-starved, meaning the conversion of X-rays to optical light limits the image quality. [In medical imaging,] you do not want to irradiate your patients with too much of the X-rays, especially for routine screening, and especially for young patients as well,” Roques-Carmes says.

      “We believe that this will open a new field of research in nanophotonics,” he adds. “You can use a lot of the existing work and research that has been done in the field of nanophotonics to improve significantly on existing materials that scintillate.”

      “The research presented in this paper is hugely significant,” says Rajiv Gupta, chief of neuroradiology at Massachusetts General Hospital and an associate professor at Harvard Medical School, who was not associated with this work. “Nearly all detectors used in the $100 billion [medical X-ray] industry are indirect detectors,” which is the type of detector the new findings apply to, he says. “Everything that I use in my clinical practice today is based on this principle. This paper improves the efficiency of this process by 10 times. If this claim is even partially true, say the improvement is two times instead of 10 times, it would be transformative for the field!”

      Soljacic says that while their experiments proved a tenfold improvement in emission could be achieved in particular systems, by further fine-tuning the design of the nanoscale patterning, “we also show that you can get up to 100 times [improvement] in certain scintillator systems, and we believe we also have a path toward making it even better,” he says.

      Soljacic points out that in other areas of nanophotonics, a field that deals with how light interacts with materials that are structured at the nanometer scale, the development of computational simulations has enabled rapid, substantial improvements, for example in the development of solar cells and LEDs. The new models this team developed for scintillating materials could facilitate similar leaps in this technology, he says.

      Nanophotonics techniques “give you the ultimate power of tailoring and enhancing the behavior of light,” Soljacic says. “But until now, this promise, this ability to do this with scintillation was unreachable because modeling the scintillation was very challenging. Now, this work for the first time opens up this field of scintillation, fully opens it, for the application of nanophotonics techniques.” More generally, the team believes that the combination of nanophotonic and scintillators might ultimately enable higher resolution, reduced X-ray dose, and energy-resolved X-ray imaging.

      This work is “very original and excellent,” says Eli Yablonovitch, a professor of Electrical Engineering and Computer Sciences at the University of California at Berkeley, who was not associated with this research. “New scintillator concepts are very important in medical imaging and in basic research.”

      Yablonovitch adds that while the concept still needs to be proven in a practical device, he says that, “After years of research on photonic crystals in optical communication and other fields, it’s long overdue that photonic crystals should be applied to scintillators, which are of great practical importance yet have been overlooked” until this work.

      The research team included Ali Ghorashi, Steven Kooi, Yi Yang, Zin Lin, Justin Beroz, Aviram Massuda, Jamison Sloan, and Nicolas Romeo at MIT; Yang Yu at Raith America, Inc.; and Ido Kaminer at Technion in Israel. The work was supported, in part, by the U.S. Army Research Office and the U.S. Army Research Laboratory through the Institute for Soldier Nanotechnologies, by the Air Force Office of Scientific Research, and by a Mathworks Engineering Fellowship. More