Artificial intelligence

Subterms

    Latest story

    125 Shares132 Views

    How artificial intelligence can help achieve a clean energy future

    by

    There is growing attention on the links between artificial intelligence and increased energy demands. But while the power-hungry data centers being built to support AI could potentially stress electricity grids, increase customer prices and service interruptions, and generally slow the transition to clean energy, the use of artificial intelligence can also help the energy transition.For example, use of AI is reducing energy consumption and associated emissions in buildings, transportation, and industrial processes. In addition, AI is helping to optimize the design and siting of new wind and solar installations and energy storage facilities.On electric power grids, using AI algorithms to control operations is helping to increase efficiency and reduce costs, integrate the growing share of renewables, and even predict when key equipment needs servicing to prevent failure and possible blackouts. AI can help grid planners schedule investments in generation, energy storage, and other infrastructure that will be needed in the future. AI is also helping researchers discover or design novel materials for nuclear reactors, batteries, and electrolyzers.Researchers at MIT and elsewhere are actively investigating aspects of those and other opportunities for AI to support the clean energy transition. At its 2025 research conference, MITEI announced the Data Center Power Forum, a targeted research effort for MITEI member companies interested in addressing the challenges of data center power demand.Controlling real-time operationsCustomers generally rely on receiving a continuous supply of electricity, and grid operators get help from AI to make that happen — while optimizing the storage and distribution of energy from renewable sources at the same time.But with more installation of solar and wind farms — both of which provide power in smaller amounts, and intermittently — and the growing threat of weather events and cyberattacks, ensuring reliability is getting more complicated. “That’s exactly where AI can come into the picture,” explains Anuradha Annaswamy, a senior research scientist in MIT’s Department of Mechanical Engineering and director of MIT’s Active-Adaptive Control Laboratory. “Essentially, you need to introduce a whole information infrastructure to supplement and complement the physical infrastructure.”The electricity grid is a complex system that requires meticulous control on time scales ranging from decades all the way down to microseconds. The challenge can be traced to the basic laws of power physics: electricity supply must equal electricity demand at every instant, or generation can be interrupted. In past decades, grid operators generally assumed that generation was fixed — they could count on how much electricity each large power plant would produce — while demand varied over time in a fairly predictable way. As a result, operators could commission specific power plants to run as needed to meet demand the next day. If some outages occurred, specially designated units would start up as needed to make up the shortfall.Today and in the future, that matching of supply and demand must still happen, even as the number of small, intermittent sources of generation grows and weather disturbances and other threats to the grid increase. AI algorithms provide a means of achieving the complex management of information needed to forecast within just a few hours which plants should run while also ensuring that the frequency, voltage, and other characteristics of the incoming power are as required for the grid to operate properly.Moreover, AI can make possible new ways of increasing supply or decreasing demand at times when supplies on the grid run short. As Annaswamy points out, the battery in your electric vehicle (EV), as well as the one charged up by solar panels or wind turbines, can — when needed — serve as a source of extra power to be fed into the grid. And given real-time price signals, EV owners can choose to shift charging from a time when demand is peaking and prices are high to a time when demand and therefore prices are both lower. In addition, new smart thermostats can be set to allow the indoor temperature to drop or rise —  a range defined by the customer — when demand on the grid is peaking. And data centers themselves can be a source of demand flexibility: selected AI calculations could be delayed as needed to smooth out peaks in demand. Thus, AI can provide many opportunities to fine-tune both supply and demand as needed.In addition, AI makes possible “predictive maintenance.” Any downtime is costly for the company and threatens shortages for the customers served. AI algorithms can collect key performance data during normal operation and, when readings veer off from that normal, the system can alert operators that something might be going wrong, giving them a chance to intervene. That capability prevents equipment failures, reduces the need for routine inspections, increases worker productivity, and extends the lifetime of key equipment.Annaswamy stresses that “figuring out how to architect this new power grid with these AI components will require many different experts to come together.” She notes that electrical engineers, computer scientists, and energy economists “will have to rub shoulders with enlightened regulators and policymakers to make sure that this is not just an academic exercise, but will actually get implemented. All the different stakeholders have to learn from each other. And you need guarantees that nothing is going to fail. You can’t have blackouts.”Using AI to help plan investments in infrastructure for the futureGrid companies constantly need to plan for expanding generation, transmission, storage, and more, and getting all the necessary infrastructure built and operating may take many years, in some cases more than a decade. So, they need to predict what infrastructure they’ll need to ensure reliability in the future. “It’s complicated because you have to forecast over a decade ahead of time what to build and where to build it,” says Deepjyoti Deka, a research scientist in MITEI.One challenge with anticipating what will be needed is predicting how the future system will operate. “That’s becoming increasingly difficult,” says Deka, because more renewables are coming online and displacing traditional generators. In the past, operators could rely on “spinning reserves,” that is, generating capacity that’s not currently in use but could come online in a matter of minutes to meet any shortfall on the system. The presence of so many intermittent generators — wind and solar — means there’s now less stability and inertia built into the grid. Adding to the complication is that those intermittent generators can be built by various vendors, and grid planners may not have access to the physics-based equations that govern the operation of each piece of equipment at sufficiently fine time scales. “So, you probably don’t know exactly how it’s going to run,” says Deka.And then there’s the weather. Determining the reliability of a proposed future energy system requires knowing what it’ll be up against in terms of weather. The future grid has to be reliable not only in everyday weather, but also during low-probability but high-risk events such as hurricanes, floods, and wildfires, all of which are becoming more and more frequent, notes Deka. AI can help by predicting such events and even tracking changes in weather patterns due to climate change.Deka points out another, less-obvious benefit of the speed of AI analysis. Any infrastructure development plan must be reviewed and approved, often by several regulatory and other bodies. Traditionally, an applicant would develop a plan, analyze its impacts, and submit the plan to one set of reviewers. After making any requested changes and repeating the analysis, the applicant would resubmit a revised version to the reviewers to see if the new version was acceptable. AI tools can speed up the required analysis so the process moves along more quickly. Planners can even reduce the number of times a proposal is rejected by using large language models to search regulatory publications and summarize what’s important for a proposed infrastructure installation.Harnessing AI to discover and exploit advanced materials needed for the energy transition“Use of AI for materials development is booming right now,” says Ju Li, MIT’s Carl Richard Soderberg Professor of Power Engineering. He notes two main directions.First, AI makes possible faster physics-based simulations at the atomic scale. The result is a better atomic-level understanding of how composition, processing, structure, and chemical reactivity relate to the performance of materials. That understanding provides design rules to help guide the development and discovery of novel materials for energy generation, storage, and conversion needed for a sustainable future energy system.And second, AI can help guide experiments in real time as they take place in the lab. Li explains: “AI assists us in choosing the best experiment to do based on our previous experiments and — based on literature searches — makes hypotheses and suggests new experiments.”He describes what happens in his own lab. Human scientists interact with a large language model, which then makes suggestions about what specific experiments to do next. The human researcher accepts or modifies the suggestion, and a robotic arm responds by setting up and performing the next step in the experimental sequence, synthesizing the material, testing the performance, and taking images of samples when appropriate. Based on a mix of literature knowledge, human intuition, and previous experimental results, AI thus coordinates active learning that balances the goals of reducing uncertainty with improving performance. And, as Li points out, “AI has read many more books and papers than any human can, and is thus naturally more interdisciplinary.”The outcome, says Li, is both better design of experiments and speeding up the “work flow.” Traditionally, the process of developing new materials has required synthesizing the precursors, making the material, testing its performance and characterizing the structure, making adjustments, and repeating the same series of steps. AI guidance speeds up that process, “helping us to design critical, cheap experiments that can give us the maximum amount of information feedback,” says Li.“Having this capability certainly will accelerate material discovery, and this may be the thing that can really help us in the clean energy transition,” he concludes. “AI [has the potential to] lubricate the material-discovery and optimization process, perhaps shortening it from decades, as in the past, to just a few years.” MITEI’s contributionsAt MIT, researchers are working on various aspects of the opportunities described above. In projects supported by MITEI, teams are using AI to better model and predict disruptions in plasma flows inside fusion reactors — a necessity in achieving practical fusion power generation. Other MITEI-supported teams are using AI-powered tools to interpret regulations, climate data, and infrastructure maps in order to achieve faster, more adaptive electric grid planning. AI-guided development of advanced materials continues, with one MITEI project using AI to optimize solar cells and thermoelectric materials.Other MITEI researchers are developing robots that can learn maintenance tasks based on human feedback, including physical intervention and verbal instructions. The goal is to reduce costs, improve safety, and accelerate the deployment of the renewable energy infrastructure. And MITEI-funded work continues on ways to reduce the energy demand of data centers, from designing more efficient computer chips and computing algorithms to rethinking the architectural design of the buildings, for example, to increase airflow so as to reduce the need for air conditioning.In addition to providing leadership and funding for many research projects, MITEI acts as a convenor, bringing together interested parties to consider common problems and potential solutions. In May 2025, MITEI’s annual spring symposium — titled “AI and energy: Peril and promise” — brought together AI and energy experts from across academia, industry, government, and nonprofit organizations to explore AI as both a problem and a potential solution for the clean energy transition. At the close of the symposium, William H. Green, director of MITEI and Hoyt C. Hottel Professor in the MIT Department of Chemical Engineering, noted, “The challenge of meeting data center energy demand and of unlocking the potential benefits of AI to the energy transition is now a research priority for MITEI.” More

    More stories

    • 150 Shares170 Views

      in Energy

      MIT Energy Initiative conference spotlights research priorities amidst a changing energy landscape

      by

      “We’re here to talk about really substantive changes, and we want you to be a participant in that,” said Desirée Plata, the School of Engineering Distinguished Professor of Climate and Energy in MIT’s Department of Civil and Environmental Engineering, at Energizing@MIT: the MIT Energy Initiative’s (MITEI) Annual Research Conference that was held on Sept. 9-10.Plata’s words resonated with the 150-plus participants from academia, industry, and government meeting in Cambridge for the conference, whose theme was “tackling emerging energy challenges.” Meeting such challenges and ultimately altering the trajectory of global climate outcomes requires partnerships, speakers agreed.“We have to be humble and open,” said Giacomo Silvestri, chair of Eniverse Ventures at Eni, in a shared keynote address. “We cannot develop innovation just focusing on ourselves and our competencies … so we need to partner with startups, venture funds, universities like MIT and other public and private institutions.” Added his Eni colleague, Annalisa Muccioli, head of research and technology, “The energy transition is a race we can win only by combining mature solutions ready to deploy, together with emerging technologies that still require acceleration and risk management.”Research targetsIn a conference that showcased a suite of research priorities MITEI has identified as central to ensuring a low-carbon energy future, participants shared both promising discoveries and strategies for advancing proven technologies in the face of shifting political winds and policy uncertainties.One panel focused on grid resiliency — a topic that has moved from the periphery to the center of energy discourse as climate-driven disruptions, cyber threats, and the integration of renewables challenge legacy systems. A dramatic case in point: the April 2025 outage in Spain and Portugal that left millions without power for eight to 15 hours. “I want to emphasize that this failure was about more than the power system,” said MITEI research scientist Pablo Duenas-Martinez. While he pinpointed technical problems with reactive power and voltage control behind the system collapse, Duenas-Martinez also called out a lack of transmission capacity with Central Europe and out-of-date operating procedures, and recommended better preparation and communication among transmission systems and utility operators.“You can’t plan for every single eventuality, which means we need to broaden the portfolio of extreme events we prepare for,” noted Jennifer Pearce, vice president at energy company Avangrid. “We are making the system smarter, stronger, and more resilient to better protect from a wide range of threats such as storms, flooding, and extreme heat events.” Pearce noted that Avangrid’s commitment to deliver safe, reliable power to its customers necessitates “meticulous emergency planning procedures.”The resiliency of the electric grid under greatly increased demand is an important motivation behind MITEI’s September 2025 launch of the Data Center Power Forum, which was also announced during the annual research conference. The forum will include research projects, webinars, and other content focused on energy supply and storage, grid design and management, infrastructure, and public and economic policy related to data centers. The forum’s members include MITEI companies that also participate in MIT’s Center for Environmental and Energy Policy Research (CEEPR).Storage and transportation: Staggering challengesMeeting climate goals to decarbonize the world by 2050 requires building around 300 terawatt-hours of storage, according to Asegun Henry, a professor in the MIT Department of Mechanical Engineering. “It’s an unbelievably enormous problem people have to wrap their minds around,” he said. Henry has been developing a high-temperature thermal energy storage system he has nicknamed “sun in a box.” His system uses liquid metal and graphite to hold electricity as heat and then convert it back to electricity, enabling storage anywhere from five to 500 hours.“At the end of the day, storage provides a service, and the type of technology that you need is a function of the service that you value the most,” said Nestor Sepulveda, commercial lead for advanced energy investments and partnerships at Google. “I don’t think there is one winner-takes-all type of market here.”Another panel explored sustainable fuels that could help decarbonize hard-to-electrify sectors like aviation, shipping, and long-haul trucking. Randall Field, MITEI’s director of research, noted that sustainably produced drop-in fuels — fuels that are largely compatible with existing engines — “could eliminate potentially trillions of dollars of cost for fleet replacement and for infrastructure build-out, while also helping us to accelerate the rate of decarbonization of the transportation sectors.”Erik G. Birkerts is the chief growth officer of LanzaJet, which produces a drop-in, high-energy-density aviation fuel derived from agricultural residue and other waste carbon sources. “The key to driving broad sustainable aviation fuel adoption is solving both the supply-side challenge through more production and the demand-side hurdle by reducing costs,” he said.“We think a good policy framework [for sustainable fuels] would be something that is technology-neutral, does not exclude any pathways to produce, is based on life cycle accounting practices, and on market mechanisms,” said Veronica L. Robertson, energy products technology portfolio manager at ExxonMobil.MITEI plans a major expansion of its research on sustainable fuels, announcing a two-year study, “The future of fuels: Pathways to sustainable transportation,” starting in early 2026. According to Field, the study will analyze and assess biofuels and e-fuels.Solutions from labs big and smallGlobal energy leaders offered glimpses of their research projects. A panel on carbon capture in power generation featured three takes on the topic: Devin Shaw, commercial director of decarbonization technologies at Shell, described post-combustion carbon capture in power plants using steam for heat recovery; Jan Marsh, a global program lead at Siemens Energy, discussed deploying novel materials to capture carbon dioxide directly from the air; and Jeffrey Goldmeer, senior director of technology strategy at GE Vernova, explained integrating carbon capture into gas-powered turbine systems.During a panel on vehicle electrification, Brian Storey, vice president of energy and materials at the Toyota Research Institute, provided an overview of Toyota’s portfolio of projects for decarbonization, including solid-state batteries, flexible manufacturing lines, and grid-forming inverters to support EV charging infrastructure.A session on MITEI seed fund projects revealed promising early-stage research inside MIT’s own labs. A new process for decarbonizing the production of ethylene was presented by Yogesh Surendranath, Donner Professor of Science in the MIT Department of Chemistry. Materials Science and Engineering assistant professor Aristide Gumyusenge also discussed the development of polymers essential for a new kind of sodium-ion battery.Shepherding bold, new technologies like these from academic labs into the real world cannot succeed without ample support and deft management. A panel on paths to commercialization featured the work of Iwnetim Abate, Chipman Career Development Professor and assistant professor in the MIT Department of Materials Science and Engineering, who has spun out a company, Addis Energy, based on a novel geothermal process for harvesting clean hydrogen and ammonia from subsurface, iron-rich rocks. Among his funders: ARPA-E and MIT’s own The Engine Ventures.The panel also highlighted the MIT Proto Ventures Program, an initiative to seize early-stage MIT ideas and unleash them as world-changing startups. “A mere 4.2 percent of all the patents that are actually prosecuted in the world are ever commercialized, which seems like a shocking number,” said Andrew Inglis, an entrepreneur working with Proto Ventures to translate geothermal discoveries into businesses. “Can’t we do this better? Let’s do this better!”Geopolitical hazardsThroughout the conference, participants often voiced concern about the impacts of competition between the United States and China. Kelly Sims Gallagher, dean of the Fletcher School at Tufts University and an expert on China’s energy landscape, delivered the sobering news in her keynote address: “U.S. competitiveness in low-carbon technologies has eroded in nearly every category,” she said. “The Chinese are winning the clean tech race.”China enjoys a 51 percent share in global wind turbine manufacture and 75 percent in solar modules. It also controls low-carbon supply chains that much of the world depends on. “China is getting so dominant that nobody can carve out a comparative advantage in anything,” said Gallagher. “China is just so big, and the scale is so huge that the Chinese can truly conquer markets and make it very hard for potential competitors to find a way in.”And for the United States, the problem is “the seesaw of energy policy,” she says. “It’s incredibly difficult for the private sector to plan and to operate, given the lack of predictability and policy here.”Nevertheless, Gallagher believes the United States still has a chance of at least regaining competitiveness, by setting up a stable, bipartisan energy policy, rebuilding domestic manufacturing and supply chains; providing consistent fiscal incentives; attracting and retaining global talent; and fostering international collaboration.The conference shone a light on one such collaboration: a China-U.S. joint venture to manufacture lithium iron phosphate batteries for commercial vehicles in the United States. The venture brings together Eve Energy, a Chinese battery technology and manufacturing company; Daimler, a global commercial vehicle manufacturer; PACCAR Inc., a U.S.-based truck manufacturer; and Accelera, the zero-emissions business of Cummins Inc. “Manufacturing batteries in the U.S. makes the supply chain more robust and reduces geopolitical risks,” said Mike Gerty, of PACCAR.While she acknowledged the obstacles confronting her colleagues in the room, Plata nevertheless concluded her remarks as a panel moderator with some optimism: “I hope you all leave this conference and look back on it in the future, saying I was in the room when they actually solved some of the challenges standing between now and the future that we all wish to manifest.” More

    • 100 Shares99 Views

      in Energy

      MIT Energy Initiative launches Data Center Power Forum

      by

      With global power demand from data centers expected to more than double by 2030, the MIT Energy Initiative (MITEI) in September launched an effort that brings together MIT researchers and industry experts to explore innovative solutions for powering the data-driven future. At its annual research conference, MITEI announced the Data Center Power Forum, a targeted research effort for MITEI member companies interested in addressing the challenges of data center power demand. The Data Center Power Forum builds on lessons from MITEI’s May 2025 symposium on the energy to power the expansion of artificial intelligence (AI) and focus panels related to data centers at the fall 2024 research conference.In the United States, data centers consumed 4 percent of the country’s electricity in 2023, with demand expected to increase to 9 percent by 2030, according to the Electric Power Research Institute. Much of the growth in demand is from the increasing use of AI, which is placing an unprecedented strain on the electric grid. This surge in demand presents a serious challenge for the technology and energy sectors, government policymakers, and everyday consumers, who may see their electric bills skyrocket as a result.“MITEI has long supported research on ways to produce more efficient and cleaner energy and to manage the electric grid. In recent years, MITEI has also funded dozens of research projects relevant to data center energy issues. Building on this history and knowledge base, MITEI’s Data Center Power Forum is convening a specialized community of industry members who have a vital stake in the sustainable growth of AI and the acceleration of solutions for powering data centers and expanding the grid,” says William H. Green, the director of MITEI and the Hoyt C. Hottel Professor of Chemical Engineering.MITEI’s mission is to advance zero- and low-carbon solutions to expand energy access and mitigate climate change. MITEI works with companies from across the energy innovation chain, including in the infrastructure, automotive, electric power, energy, natural resources, and insurance sectors. MITEI member companies have expressed strong interest in the Data Center Power Forum and are committing to support focused research on a wide range of energy issues associated with data center expansion, Green says.MITEI’s Data Center Power Forum will provide its member companies with reliable insights into energy supply, grid load operations and management, the built environment, and electricity market design and regulatory policy for data centers. The forum complements MIT’s deep expertise in adjacent topics such as low-power processors, efficient algorithms, task-specific AI, photonic devices, quantum computing, and the societal consequences of data center expansion. As part of the forum, MITEI’s Future Energy Systems Center is funding projects relevant to data center energy in its upcoming proposal cycles. MITEI Research Scientist Deep Deka has been named the program manager for the forum.“Figuring out how to meet the power demands of data centers is a complicated challenge. Our research is coming at this from multiple directions, from looking at ways to expand transmission capacity within the electrical grid in order to bring power to where it is needed, to ensuring the quality of electrical service for existing users is not diminished when new data centers come online, and to shifting computing tasks to times and places when and where energy is available on the grid,” said Deka.MITEI currently sponsors substantial research related to data center energy topics across several MIT departments. The existing research portfolio includes more than a dozen projects related to data centers, including low- or zero-carbon solutions for energy supply and infrastructure, electrical grid management, and electricity market policy. MIT researchers funded through MITEI’s industry consortium are also designing more energy-efficient power electronics and processors and investigating behind-the-meter low-/no-carbon power plants and energy storage. MITEI-supported experts are studying how to use AI to optimize electrical distribution and the siting of data centers and conducting techno-economic analyses of data center power schemes. MITEI’s consortium projects are also bringing fresh perspectives to data center cooling challenges and considering policy approaches to balance the interests of shareholders. By drawing together industry stakeholders from across the AI and grid value chain, the Data Center Power Forum enables a richer dialog about solutions to power, grid, and carbon management problems in a noncommercial and collaborative setting.“The opportunity to meet and to hold discussions on key data center challenges with other forum members from different sectors, as well as with MIT faculty members and research scientists, is a unique benefit of this MITEI-led effort,” Green says.MITEI addressed the issue of data center power needs with its company members during its fall 2024 Annual Research Conference with a panel session titled, “The extreme challenge of powering data centers in a decarbonized way.” MITEI Director of Research Randall Field led a discussion with representatives from large technology companies Google and Microsoft, known as “hyperscalers,” as well as Madrid-based infrastructure developer Ferrovial S.E. and utility company Exelon Corp. Another conference session addressed the related topic, “Energy storage and grid expansion.” This past spring, MITEI focused its annual Spring Symposium on data centers, hosting faculty members and researchers from MIT and other universities, business leaders, and a representative of the Federal Energy Regulatory Commission for a full day of sessions on the topic, “AI and energy: Peril and promise.”  More

    • 150 Shares169 Views

      in Environment

      3 Questions: How AI is helping us monitor and support vulnerable ecosystems

      by

      A recent study from Oregon State University estimated that more than 3,500 animal species are at risk of extinction because of factors including habitat alterations, natural resources being overexploited, and climate change.To better understand these changes and protect vulnerable wildlife, conservationists like MIT PhD student and Computer Science and Artificial Intelligence Laboratory (CSAIL) researcher Justin Kay are developing computer vision algorithms that carefully monitor animal populations. A member of the lab of MIT Department of Electrical Engineering and Computer Science assistant professor and CSAIL principal investigator Sara Beery, Kay is currently working on tracking salmon in the Pacific Northwest, where they provide crucial nutrients to predators like birds and bears, while managing the population of prey, like bugs.With all that wildlife data, though, researchers have lots of information to sort through and many AI models to choose from to analyze it all. Kay and his colleagues at CSAIL and the University of Massachusetts Amherst are developing AI methods that make this data-crunching process much more efficient, including a new approach called “consensus-driven active model selection” (or “CODA”) that helps conservationists choose which AI model to use. Their work was named a Highlight Paper at the International Conference on Computer Vision (ICCV) in October.That research was supported, in part, by the National Science Foundation, Natural Sciences and Engineering Research Council of Canada, and Abdul Latif Jameel Water and Food Systems Lab (J-WAFS). Here, Kay discusses this project, among other conservation efforts.Q: In your paper, you pose the question of which AI models will perform the best on a particular dataset. With as many as 1.9 million pre-trained models available in the HuggingFace Models repository alone, how does CODA help us address that challenge?A: Until recently, using AI for data analysis has typically meant training your own model. This requires significant effort to collect and annotate a representative training dataset, as well as iteratively train and validate models. You also need a certain technical skill set to run and modify AI training code. The way people interact with AI is changing, though — in particular, there are now millions of publicly available pre-trained models that can perform a variety of predictive tasks very well. This potentially enables people to use AI to analyze their data without developing their own model, simply by downloading an existing model with the capabilities they need. But this poses a new challenge: Which model, of the millions available, should they use to analyze their data? Typically, answering this model selection question also requires you to spend a lot of time collecting and annotating a large dataset, albeit for testing models rather than training them. This is especially true for real applications where user needs are specific, data distributions are imbalanced and constantly changing, and model performance may be inconsistent across samples. Our goal with CODA was to substantially reduce this effort. We do this by making the data annotation process “active.” Instead of requiring users to bulk-annotate a large test dataset all at once, in active model selection we make the process interactive, guiding users to annotate the most informative data points in their raw data. This is remarkably effective, often requiring users to annotate as few as 25 examples to identify the best model from their set of candidates. We’re very excited about CODA offering a new perspective on how to best utilize human effort in the development and deployment of machine-learning (ML) systems. As AI models become more commonplace, our work emphasizes the value of focusing effort on robust evaluation pipelines, rather than solely on training.Q: You applied the CODA method to classifying wildlife in images. Why did it perform so well, and what role can systems like this have in monitoring ecosystems in the future?A: One key insight was that when considering a collection of candidate AI models, the consensus of all of their predictions is more informative than any individual model’s predictions. This can be seen as a sort of “wisdom of the crowd:” On average, pooling the votes of all models gives you a decent prior over what the labels of individual data points in your raw dataset should be. Our approach with CODA is based on estimating a “confusion matrix” for each AI model — given the true label for some data point is class X, what is the probability that an individual model predicts class X, Y, or Z? This creates informative dependencies between all of the candidate models, the categories you want to label, and the unlabeled points in your dataset.Consider an example application where you are a wildlife ecologist who has just collected a dataset containing potentially hundreds of thousands of images from cameras deployed in the wild. You want to know what species are in these images, a time-consuming task that computer vision classifiers can help automate. You are trying to decide which species classification model to run on your data. If you have labeled 50 images of tigers so far, and some model has performed well on those 50 images, you can be pretty confident it will perform well on the remainder of the (currently unlabeled) images of tigers in your raw dataset as well. You also know that when that model predicts some image contains a tiger, it is likely to be correct, and therefore that any model that predicts a different label for that image is more likely to be wrong. You can use all these interdependencies to construct probabilistic estimates of each model’s confusion matrix, as well as a probability distribution over which model has the highest accuracy on the overall dataset. These design choices allow us to make more informed choices over which data points to label and ultimately are the reason why CODA performs model selection much more efficiently than past work.There are also a lot of exciting possibilities for building on top of our work. We think there may be even better ways of constructing informative priors for model selection based on domain expertise — for instance, if it is already known that one model performs exceptionally well on some subset of classes or poorly on others. There are also opportunities to extend the framework to support more complex machine-learning tasks and more sophisticated probabilistic models of performance. We hope our work can provide inspiration and a starting point for other researchers to keep pushing the state of the art.Q: You work in the Beerylab, led by Sara Beery, where researchers are combining the pattern-recognition capabilities of machine-learning algorithms with computer vision technology to monitor wildlife. What are some other ways your team is tracking and analyzing the natural world, beyond CODA?A: The lab is a really exciting place to work, and new projects are emerging all the time. We have ongoing projects monitoring coral reefs with drones, re-identifying individual elephants over time, and fusing multi-modal Earth observation data from satellites and in-situ cameras, just to name a few. Broadly, we look at emerging technologies for biodiversity monitoring and try to understand where the data analysis bottlenecks are, and develop new computer vision and machine-learning approaches that address those problems in a widely applicable way. It’s an exciting way of approaching problems that sort of targets the “meta-questions” underlying particular data challenges we face. The computer vision algorithms I’ve worked on that count migrating salmon in underwater sonar video are examples of that work. We often deal with shifting data distributions, even as we try to construct the most diverse training datasets we can. We always encounter something new when we deploy a new camera, and this tends to degrade the performance of computer vision algorithms. This is one instance of a general problem in machine learning called domain adaptation, but when we tried to apply existing domain adaptation algorithms to our fisheries data we realized there were serious limitations in how existing algorithms were trained and evaluated. We were able to develop a new domain adaptation framework, published earlier this year in Transactions on Machine Learning Research, that addressed these limitations and led to advancements in fish counting, and even self-driving and spacecraft analysis.One line of work that I’m particularly excited about is understanding how to better develop and analyze the performance of predictive ML algorithms in the context of what they are actually used for. Usually, the outputs from some computer vision algorithm — say, bounding boxes around animals in images — are not actually the thing that people care about, but rather a means to an end to answer a larger problem — say, what species live here, and how is that changing over time? We have been working on methods to analyze predictive performance in this context and reconsider the ways that we input human expertise into ML systems with this in mind. CODA was one example of this, where we showed that we could actually consider the ML models themselves as fixed and build a statistical framework to understand their performance very efficiently. We have been working recently on similar integrated analyses combining ML predictions with multi-stage prediction pipelines, as well as ecological statistical models. The natural world is changing at unprecedented rates and scales, and being able to quickly move from scientific hypotheses or management questions to data-driven answers is more important than ever for protecting ecosystems and the communities that depend on them. Advancements in AI can play an important role, but we need to think critically about the ways that we design, train, and evaluate algorithms in the context of these very real challenges. More

    • 75 Shares169 Views

      in Energy

      The brain power behind sustainable AI

      by

      How can you use science to build a better gingerbread house?That was something Miranda Schwacke spent a lot of time thinking about. The MIT graduate student in the Department of Materials Science and Engineering (DMSE) is part of Kitchen Matters, a group of grad students who use food and kitchen tools to explain scientific concepts through short videos and outreach events. Past topics included why chocolate “seizes,” or becomes difficult to work with when melting (spoiler: water gets in), and how to make isomalt, the sugar glass that stunt performers jump through in action movies.Two years ago, when the group was making a video on how to build a structurally sound gingerbread house, Schwacke scoured cookbooks for a variable that would produce the most dramatic difference in the cookies.“I was reading about what determines the texture of cookies, and then tried several recipes in my kitchen until I got two gingerbread recipes that I was happy with,” Schwacke says.She focused on butter, which contains water that turns to steam at high baking temperatures, creating air pockets in cookies. Schwacke predicted that decreasing the amount of butter would yield denser gingerbread, strong enough to hold together as a house.“This hypothesis is an example of how changing the structure can influence the properties and performance of material,” Schwacke said in the eight-minute video.That same curiosity about materials properties and performance drives her research on the high energy cost of computing, especially for artificial intelligence. Schwacke develops new materials and devices for neuromorphic computing, which mimics the brain by processing and storing information in the same place. She studies electrochemical ionic synapses — tiny devices that can be “tuned” to adjust conductivity, much like neurons strengthening or weakening connections in the brain.“If you look at AI in particular — to train these really large models — that consumes a lot of energy. And if you compare that to the amount of energy that we consume as humans when we’re learning things, the brain consumes a lot less energy,” Schwacke says. “That’s what led to this idea to find more brain-inspired, energy-efficient ways of doing AI.”Her advisor, Bilge Yildiz, underscores the point: One reason the brain is so efficient is that data doesn’t need to be moved back and forth.“In the brain, the connections between our neurons, called synapses, are where we process information. Signal transmission is there. It is processed, programmed, and also stored in the same place,” says Yildiz, the Breene M. Kerr (1951) Professor in the Department of Nuclear Science and Engineering and DMSE. Schwacke’s devices aim to replicate that efficiency.Scientific rootsThe daughter of a marine biologist mom and an electrical engineer dad, Schwacke was immersed in science from a young age. Science was “always a part of how I understood the world.”“I was obsessed with dinosaurs. I wanted to be a paleontologist when I grew up,” she says. But her interests broadened. At her middle school in Charleston, South Carolina, she joined a FIRST Lego League robotics competition, building robots to complete tasks like pushing or pulling objects. “My parents, my dad especially, got very involved in the school team and helping us design and build our little robot for the competition.”Her mother, meanwhile, studied how dolphin populations are affected by pollution for the National Oceanic and Atmospheric Administration. That had a lasting impact.“That was an example of how science can be used to understand the world, and also to figure out how we can improve the world,” Schwacke says. “And that’s what I’ve always wanted to do with science.”Her interest in materials science came later, in her high school magnet program. There, she was introduced to the interdisciplinary subject, a blend of physics, chemistry, and engineering that studies the structure and properties of materials and uses that knowledge to design new ones.“I always liked that it goes from this very basic science, where we’re studying how atoms are ordering, all the way up to these solid materials that we interact with in our everyday lives — and how that gives them their properties that we can see and play with,” Schwacke says.As a senior, she participated in a research program with a thesis project on dye-sensitized solar cells, a low-cost, lightweight solar technology that uses dye molecules to absorb light and generate electricity.“What drove me was really understanding, this is how we go from light to energy that we can use — and also seeing how this could help us with having more renewable energy sources,” Schwacke says.After high school, she headed across the country to Caltech. “I wanted to try a totally new place,” she says, where she studied materials science, including nanostructured materials thousands of times thinner than a human hair. She focused on materials properties and microstructure — the tiny internal structure that governs how materials behave — which led her to electrochemical systems like batteries and fuel cells.AI energy challengeAt MIT, she continued exploring energy technologies. She met Yildiz during a Zoom meeting in her first year of graduate school, in fall 2020, when the campus was still operating under strict Covid-19 protocols. Yildiz’s lab studies how charged atoms, or ions, move through materials in technologies like fuel cells, batteries, and electrolyzers.The lab’s research into brain-inspired computing fired Schwacke’s imagination, but she was equally drawn to Yildiz’s way of talking about science.“It wasn’t based on jargon and emphasized a very basic understanding of what was going on — that ions are going here, and electrons are going here — to understand fundamentally what’s happening in the system,” Schwacke says.That mindset shaped her approach to research. Her early projects focused on the properties these devices need to work well — fast operation, low energy use, and compatibility with semiconductor technology — and on using magnesium ions instead of hydrogen, which can escape into the environment and make devices unstable.Her current project, the focus of her PhD thesis, centers on understanding how the insertion of magnesium ions into tungsten oxide, a metal oxide whose electrical properties can be precisely tuned, changes its electrical resistance. In these devices, tungsten oxide serves as a channel layer, where resistance controls signal strength, much like synapses regulate signals in the brain.“I am trying to understand exactly how these devices change the channel conductance,” Schwacke says.Schwacke’s research was recognized with a MathWorks Fellowship from the School of Engineering in 2023 and 2024. The fellowship supports graduate students who leverage tools like MATLAB or Simulink in their work; Schwacke applied MATLAB for critical data analysis and visualization.Yildiz describes Schwacke’s research as a novel step toward solving one of AI’s biggest challenges.“This is electrochemistry for brain-inspired computing,” Yildiz says. “It’s a new context for electrochemistry, but also with an energy implication, because the energy consumption of computing is unsustainably increasing. We have to find new ways of doing computing with much lower energy, and this is one way that can help us move in that direction.”Like any pioneering work, it comes with challenges, especially in bridging the concepts between electrochemistry and semiconductor physics.“Our group comes from a solid-state chemistry background, and when we started this work looking into magnesium, no one had used magnesium in these kinds of devices before,” Schwacke says. “So we were looking at the magnesium battery literature for inspiration and different materials and strategies we could use. When I started this, I wasn’t just learning the language and norms for one field — I was trying to learn it for two fields, and also translate between the two.”She also grapples with a challenge familiar to all scientists: how to make sense of messy data.“The main challenge is being able to take my data and know that I’m interpreting it in a way that’s correct, and that I understand what it actually means,” Schwacke says.She overcomes hurdles by collaborating closely with colleagues across fields, including neuroscience and electrical engineering, and sometimes by just making small changes to her experiments and watching what happens next.Community mattersSchwacke is not just active in the lab. In Kitchen Matters, she and her fellow DMSE grad students set up booths at local events like the Cambridge Science Fair and Steam It Up, an after-school program with hands-on activities for kids.“We did ‘pHun with Food’ with ‘fun’ spelled with a pH, so we had cabbage juice as a pH indicator,” Schwacke says. “We let the kids test the pH of lemon juice and vinegar and dish soap, and they had a lot of fun mixing the different liquids and seeing all the different colors.”She has also served as the social chair and treasurer for DMSE’s graduate student group, the Graduate Materials Council. As an undergraduate at Caltech, she led workshops in science and technology for Robogals, a student-run group that encourages young women to pursue careers in science, and assisted students in applying for the school’s Summer Undergraduate Research Fellowships.For Schwacke, these experiences sharpened her ability to explain science to different audiences, a skill she sees as vital whether she’s presenting at a kids’ fair or at a research conference.“I always think, where is my audience starting from, and what do I need to explain before I can get into what I’m doing so that it’ll all make sense to them?” she says.Schwacke sees the ability to communicate as central to building community, which she considers an important part of doing research. “It helps with spreading ideas. It always helps to get a new perspective on what you’re working on,” she says. “I also think it keeps us sane during our PhD.”Yildiz sees Schwacke’s community involvement as an important part of her resume. “She’s doing all these activities to motivate the broader community to do research, to be interested in science, to pursue science and technology, but that ability will help her also progress in her own research and academic endeavors.”After her PhD, Schwacke wants to take that ability to communicate with her to academia, where she’d like to inspire the next generation of scientists and engineers. Yildiz has no doubt she’ll thrive.“I think she’s a perfect fit,” Yildiz says. “She’s brilliant, but brilliance by itself is not enough. She’s persistent, resilient. You really need those on top of that.” More

    • 113 Shares169 Views

      in Security

      Optimizing food subsidies: Applying digital platforms to maximize nutrition

      by

      Oct. 16 is World Food Day, a global campaign to celebrate the founding of the Food and Agriculture Organization 80 years ago, and to work toward a healthy, sustainable, food-secure future. More than 670 million people in the world are facing hunger. Millions of others are facing rising obesity rates and struggle to get healthy food for proper nutrition. World Food Day calls on not only world governments, but business, academia, the media, and even the youth to take action to promote resilient food systems and combat hunger. This year, the Abdul Latif Jameel Water and Food Systems Laboratory (J-WAFS) is spotlighting an MIT researcher who is working toward this goal by studying food and water systems in the Global South.J-WAFS seed grants provide funding to early-stage research projects that are unique to prior work. In an 11th round of seed grant funding in 2025, 10 MIT faculty members received support to carry out their cutting-edge water and food research. Ali Aouad PhD ’17, assistant professor of operations management at the MIT Sloan School of Management, was one of those grantees. “I had searched before joining MIT what kind of research centers and initiatives were available that tried to coalesce research on food systems,” Aouad says. “And so, I was very excited about J-WAFS.” Aouad gathered more information about J-WAFS at the new faculty orientation session in August 2024, where he spoke to J-WAFS staff and learned about the program’s grant opportunities for water and food research. Later that fall semester, he attended a few J-WAFS seminars on agricultural economics and water resource management. That’s when Aouad knew that his project was perfectly aligned with the J-WAFS mission of securing humankind’s water and food.Aouad’s seed project focuses on food subsidies. With a background in operations research and an interest in digital platforms, much of his work has centered on aligning supply-side operations with heterogeneous customer preferences. Past projects include ones on retail and matching systems. “I started thinking that these types of demand-driven approaches may be also very relevant to important social challenges, particularly as they relate to food security,” Aouad says. Before starting his PhD at MIT, Aouad worked on projects that looked at subsidies for smallholder farmers in low- and middle-income countries. “I think in the back of my mind, I’ve always been fascinated by trying to solve these issues,” he noted.His seed grant project, Optimal subsidy design: Application to food assistance programs, aims to leverage data on preferences and purchasing habits from local grocery stores in India to inform food assistance policy and optimize the design of subsidies. Typical data collection systems, like point-of-sales, are not as readily available in India’s local groceries, making this type of data hard to come by for low-income individuals. “Mom-and-pop stores are extremely important last-mile operators when it comes to nutrition,” he explains. For this project, the research team gave local grocers point-of-sale scanners to track purchasing habits. “We aim to develop an algorithm that converts these transactions into some sort of ‘revelation’ of the individuals’ latent preferences,” says Aouad. “As such, we can model and optimize the food assistance programs — how much variety and flexibility is offered, taking into account the expected demand uptake.” He continues, “now, of course, our ability to answer detailed design questions [across various products and prices] depends on the quality of our inference from  the data, and so this is where we need more sophisticated and robust algorithms.”Following the data collection and model development, the ultimate goal of this research is to inform policy surrounding food assistance programs through an “optimization approach.” Aouad describes the complexities of using optimization to guide policy. “Policies are often informed by domain expertise, legacy systems, or political deliberation. A lot of researchers build rigorous evidence to inform food policy, but it’s fair to say that the kind of approach that I’m proposing in this research is not something that is commonly used. I see an opportunity for bringing a new approach and methodological tradition to a problem that has been central for policy for many decades.” The overall health of consumers is the reason food assistance programs exist, yet measuring long-term nutritional impacts and shifts in purchase behavior is difficult. In past research, Aouad notes that the short-term effects of food assistance interventions can be significant. However, these effects are often short-lived. “This is a fascinating question that I don’t think we will be able to address within the space of interventions that we will be considering. However, I think it is something I would like to capture in the research, and maybe develop hypotheses for future work around how we can shift nutrition-related behaviors in the long run.”While his project develops a new methodology to calibrate food assistance programs, large-scale applications are not promised. “A lot of what drives subsidy mechanisms and food assistance programs is also, quite frankly, how easy it is and how cost-effective it is to implement these policies in the first place,” comments Aouad. Cost and infrastructure barriers are unavoidable to this kind of policy research, as well as sustaining these programs. Aouad’s effort will provide insights into customer preferences and subsidy optimization in a pilot setup, but replicating this approach on a real scale may be costly. Aouad hopes to be able to gather proxy information from customers that would both feed into the model and provide insight into a more cost-effective way to collect data for large-scale implementation.There is still much work to be done to ensure food security for all, whether it’s advances in agriculture, food-assistance programs, or ways to boost adequate nutrition. As the 2026 seed grant deadline approaches, J-WAFS will continue its mission of supporting MIT faculty as they pursue innovative projects that have practical and real impacts on water and food system challenges. More

    • 113 Shares169 Views

      in Energy

      Fighting for the health of the planet with AI

      by

      For Priya Donti, childhood trips to India were more than an opportunity to visit extended family. The biennial journeys activated in her a motivation that continues to shape her research and her teaching.Contrasting her family home in Massachusetts, Donti — now the Silverman Family Career Development Professor in the Department of Electrical Engineering and Computer Science (EECS), a shared position between the MIT Schwarzman College of Computing and EECS, and a principal investigator at the MIT Laboratory for Information and Decision Systems (LIDS) — was struck by the disparities in how people live.“It was very clear to me the extent to which inequity is a rampant issue around the world,” Donti says. “From a young age, I knew that I definitely wanted to address that issue.”That motivation was further stoked by a high school biology teacher, who focused his class on climate and sustainability.“We learned that climate change, this huge, important issue, would exacerbate inequity,” Donti says. “That really stuck with me and put a fire in my belly.”So, when Donti enrolled at Harvey Mudd College, she thought she would direct her energy toward the study of chemistry or materials science to create next-generation solar panels.Those plans, however, were jilted. Donti “fell in love” with computer science, and then discovered work by researchers in the United Kingdom who were arguing that artificial intelligence and machine learning would be essential to help integrate renewables into power grids.“It was the first time I’d seen those two interests brought together,” she says. “I got hooked and have been working on that topic ever since.”Pursuing a PhD at Carnegie Mellon University, Donti was able to design her degree to include computer science and public policy. In her research, she explored the need for fundamental algorithms and tools that could manage, at scale, power grids relying heavily on renewables.“I wanted to have a hand in developing those algorithms and tool kits by creating new machine learning techniques grounded in computer science,” she says. “But I wanted to make sure that the way I was doing the work was grounded both in the actual energy systems domain and working with people in that domain” to provide what was actually needed.While Donti was working on her PhD, she co-founded a nonprofit called Climate Change AI. Her objective, she says, was to help the community of people involved in climate and sustainability — “be they computer scientists, academics, practitioners, or policymakers” — to come together and access resources, connection, and education “to help them along that journey.”“In the climate space,” she says, “you need experts in particular climate change-related sectors, experts in different technical and social science tool kits, problem owners, affected users, policymakers who know the regulations — all of those — to have on-the-ground scalable impact.”When Donti came to MIT in September 2023, it was not surprising that she was drawn by its initiatives directing the application of computer science toward society’s biggest problems, especially the current threat to the health of the planet.“We’re really thinking about where technology has a much longer-horizon impact and how technology, society, and policy all have to work together,” Donti says. “Technology is not just one-and-done and monetizable in the context of a year.”Her work uses deep learning models to incorporate the physics and hard constraints of electric power systems that employ renewables for better forecasting, optimization, and control.“Machine learning is already really widely used for things like solar power forecasting, which is a prerequisite to managing and balancing power grids,” she says. “My focus is, how do you improve the algorithms for actually balancing power grids in the face of a range of time-varying renewables?”Among Donti’s breakthroughs is a promising solution for power grid operators to be able to optimize for cost, taking into account the actual physical realities of the grid, rather than relying on approximations. While the solution is not yet deployed, it appears to work 10 times faster, and far more cheaply, than previous technologies, and has attracted the attention of grid operators.Another technology she is developing works to provide data that can be used in training machine learning systems for power system optimization. In general, much data related to the systems is private, either because it is proprietary or because of security concerns. Donti and her research group are working to create synthetic data and benchmarks that, Donti says, “can help to expose some of the underlying problems” in making power systems more efficient.“The question is,” Donti says, “can we bring our datasets to a point such that they are just hard enough to drive progress?”For her efforts, Donti has been awarded the U.S. Department of Energy Computational Science Graduate Fellowship and the NSF Graduate Research Fellowship. She was recognized as part of MIT Technology Review’s 2021 list of “35 Innovators Under 35” and Vox’s 2023 “Future Perfect 50.”Next spring, Donti will co-teach a class called AI for Climate Action with Sara Beery, EECS assistant professor, whose focus is AI for biodiversity and ecosystems, and Abigail Bodner, assistant professor in the departments of EECS and Earth, Atmospheric and Planetary Sciences, whose focus is AI for climate and Earth science.“We’re all super-excited about it,” Donti says.Coming to MIT, Donti says, “I knew that there would be an ecosystem of people who really cared, not just about success metrics like publications and citation counts, but about the impact of our work on society.” More

    • 138 Shares129 Views

      in Energy

      New prediction model could improve the reliability of fusion power plants

      by

      Tokamaks are machines that are meant to hold and harness the power of the sun. These fusion machines use powerful magnets to contain a plasma hotter than the sun’s core and push the plasma’s atoms to fuse and release energy. If tokamaks can operate safely and efficiently, the machines could one day provide clean and limitless fusion energy.Today, there are a number of experimental tokamaks in operation around the world, with more underway. Most are small-scale research machines built to investigate how the devices can spin up plasma and harness its energy. One of the challenges that tokamaks face is how to safely and reliably turn off a plasma current that is circulating at speeds of up to 100 kilometers per second, at temperatures of over 100 million degrees Celsius.Such “rampdowns” are necessary when a plasma becomes unstable. To prevent the plasma from further disrupting and potentially damaging the device’s interior, operators ramp down the plasma current. But occasionally the rampdown itself can destabilize the plasma. In some machines, rampdowns have caused scrapes and scarring to the tokamak’s interior — minor damage that still requires considerable time and resources to repair.Now, scientists at MIT have developed a method to predict how plasma in a tokamak will behave during a rampdown. The team combined machine-learning tools with a physics-based model of plasma dynamics to simulate a plasma’s behavior and any instabilities that may arise as the plasma is ramped down and turned off. The researchers trained and tested the new model on plasma data from an experimental tokamak in Switzerland. They found the method quickly learned how plasma would evolve as it was tuned down in different ways. What’s more, the method achieved a high level of accuracy using a relatively small amount of data. This training efficiency is promising, given that each experimental run of a tokamak is expensive and quality data is limited as a result.The new model, which the team highlights this week in an open-access Nature Communications paper, could improve the safety and reliability of future fusion power plants.“For fusion to be a useful energy source it’s going to have to be reliable,” says lead author Allen Wang, a graduate student in aeronautics and astronautics and a member of the Disruption Group at MIT’s Plasma Science and Fusion Center (PSFC). “To be reliable, we need to get good at managing our plasmas.”The study’s MIT co-authors include PSFC Principal Research Scientist and Disruptions Group leader Cristina Rea, and members of the Laboratory for Information and Decision Systems (LIDS) Oswin So, Charles Dawson, and Professor Chuchu Fan, along with Mark (Dan) Boyer of Commonwealth Fusion Systems and collaborators from the Swiss Plasma Center in Switzerland.“A delicate balance”Tokamaks are experimental fusion devices that were first built in the Soviet Union in the 1950s. The device gets its name from a Russian acronym that translates to a “toroidal chamber with magnetic coils.” Just as its name describes, a tokamak is toroidal, or donut-shaped, and uses powerful magnets to contain and spin up a gas to temperatures and energies high enough that atoms in the resulting plasma can fuse and release energy.Today, tokamak experiments are relatively low-energy in scale, with few approaching the size and output needed to generate safe, reliable, usable energy. Disruptions in experimental, low-energy tokamaks are generally not an issue. But as fusion machines scale up to grid-scale dimensions, controlling much higher-energy plasmas at all phases will be paramount to maintaining a machine’s safe and efficient operation.“Uncontrolled plasma terminations, even during rampdown, can generate intense heat fluxes damaging the internal walls,” Wang notes. “Quite often, especially with the high-performance plasmas, rampdowns actually can push the plasma closer to some instability limits. So, it’s a delicate balance. And there’s a lot of focus now on how to manage instabilities so that we can routinely and reliably take these plasmas and safely power them down. And there are relatively few studies done on how to do that well.”Bringing down the pulseWang and his colleagues developed a model to predict how a plasma will behave during tokamak rampdown. While they could have simply applied machine-learning tools such as a neural network to learn signs of instabilities in plasma data, “you would need an ungodly amount of data” for such tools to discern the very subtle and ephemeral changes in extremely high-temperature, high-energy plasmas, Wang says.Instead, the researchers paired a neural network with an existing model that simulates plasma dynamics according to the fundamental rules of physics. With this combination of machine learning and a physics-based plasma simulation, the team found that only a couple hundred pulses at low performance, and a small handful of pulses at high performance, were sufficient to train and validate the new model.The data they used for the new study came from the TCV, the Swiss “variable configuration tokamak” operated by the Swiss Plasma Center at EPFL (the Swiss Federal Institute of Technology Lausanne). The TCV is a small experimental fusion experimental device that is used for research purposes, often as test bed for next-generation device solutions. Wang used the data from several hundred TCV plasma pulses that included properties of the plasma such as its temperature and energies during each pulse’s ramp-up, run, and ramp-down. He trained the new model on this data, then tested it and found it was able to accurately predict the plasma’s evolution given the initial conditions of a particular tokamak run.The researchers also developed an algorithm to translate the model’s predictions into practical “trajectories,” or plasma-managing instructions that a tokamak controller can automatically carry out to for instance adjust the magnets or temperature maintain the plasma’s stability. They implemented the algorithm on several TCV runs and found that it produced trajectories that safely ramped down a plasma pulse, in some cases faster and without disruptions compared to runs without the new method.“At some point the plasma will always go away, but we call it a disruption when the plasma goes away at high energy. Here, we ramped the energy down to nothing,” Wang notes. “We did it a number of times. And we did things much better across the board. So, we had statistical confidence that we made things better.”The work was supported in part by Commonwealth Fusion Systems (CFS), an MIT spinout that intends to build the world’s first compact, grid-scale fusion power plant. The company is developing a demo tokamak, SPARC, designed to produce net-energy plasma, meaning that it should generate more energy than it takes to heat up the plasma. Wang and his colleagues are working with CFS on ways that the new prediction model and tools like it can better predict plasma behavior and prevent costly disruptions to enable safe and reliable fusion power.“We’re trying to tackle the science questions to make fusion routinely useful,” Wang says. “What we’ve done here is the start of what is still a long journey. But I think we’ve made some nice progress.”Additional support for the research came from the framework of the EUROfusion Consortium, via the Euratom Research and Training Program and funded by the Swiss State Secretariat for Education, Research, and Innovation. More

    • 150 Shares169 Views

      in Energy

      Lincoln Lab unveils the most powerful AI supercomputer at any US university

      by

      The new TX-Generative AI Next (TX-GAIN) computing system at the Lincoln Laboratory Supercomputing Center  (LLSC) is the most powerful AI supercomputer at any U.S. university. With its recent ranking from  TOP500, which biannually publishes a list of the top supercomputers in various categories, TX-GAIN joins the ranks of other powerful systems at the LLSC, all supporting research and development at Lincoln Laboratory and across the MIT campus. “TX-GAIN will enable our researchers to achieve scientific and engineering breakthroughs. The system will play a large role in supporting generative AI, physical simulation, and data analysis across all research areas,” says Lincoln Laboratory Fellow Jeremy Kepner, who heads the LLSC. The LLSC is a key resource for accelerating innovation at Lincoln Laboratory. Thousands of researchers tap into the LLSC to analyze data, train models, and run simulations for federally funded research projects. The supercomputers have been used, for example, to simulate billions of aircraft encounters to develop collision-avoidance systems for the Federal Aviation Administration, and to train models in the complex tasks of autonomous navigation for the Department of Defense. Over the years, LLSC capabilities have been essential to numerous award-winning technologies, including those that have improved  airline safety,  prevented the spread of new diseases, and  aided in hurricane responses. As its name suggests, TX-GAIN is especially equipped for developing and applying generative AI. Whereas traditional AI focuses on categorization tasks, like identifying whether a photo depicts a dog or cat, generative AI produces entirely new outputs. Kepner describes it as a mathematical combination of interpolation (filling in the gaps between known data points) and extrapolation (extending data beyond known points). Today, generative AI is widely known for its use of large language models to create human-like responses to user prompts. At Lincoln Laboratory, teams are applying generative AI to various domains beyond large language models. They are using the technology, for instance, to evaluate radar signatures, supplement weather data where coverage is missing, root out anomalies in network traffic, and explore chemical interactions to design new medicines and materials.To enable such intense computations, TX-GAIN is powered by more than 600 NVIDIA graphics processing unit accelerators specially designed for AI operations, in addition to traditional high-performance computing hardware. With a peak performance of two AI exaflops (two quintillion floating-point operations per second), TX-GAIN is the top AI system at a university, and in the Northeast. Since TX-GAIN came online this summer, researchers have taken notice. “TX-GAIN is allowing us to model not only significantly more protein interactions than ever before, but also much larger proteins with more atoms. This new computational capability is a game-changer for protein characterization efforts in biological defense,” says Rafael Jaimes, a researcher in Lincoln Laboratory’s Counter–Weapons of Mass Destruction Systems Group. The LLSC’s focus on interactive supercomputing makes it especially useful to researchers. For years, the LLSC has pioneered software that lets users access its powerful systems without needing to be experts in configuring algorithms for parallel processing.  “The LLSC has always tried to make supercomputing feel like working on your laptop,” Kepner says. “The amount of data and the sophistication of analysis methods needed to be competitive today are well beyond what can be done on a laptop. But with our user-friendly approach, people can run their model and get answers quickly from their workspace.”Beyond supporting programs solely at Lincoln Laboratory, TX-GAIN is enhancing research collaborations with MIT’s campus. Such collaborations include the Haystack Observatory, Center for Quantum Engineering, Beaver Works, and Department of Air Force–MIT AI Accelerator. The latter initiative is rapidly prototyping, scaling, and applying AI technologies for the U.S. Air Force and Space Force, optimizing flight scheduling for global operations as one fielded example.The LLSC systems are housed in an energy-efficient data center and facility in Holyoke, Massachusetts. Research staff in the LLSC are also tackling the immense energy needs of AI and leading research into various power-reduction methods. One software tool they developed can reduce the energy of training an AI model by as much as 80 percent.”The LLSC provides the capabilities needed to do leading-edge research, while in a cost-effective and energy-efficient manner,” Kepner says.All of the supercomputers at the LLSC use the “TX” nomenclature in homage to Lincoln Laboratory’s Transistorized Experimental Computer Zero (TX-0) of 1956. TX-0 was one of the world’s first transistor-based machines, and its 1958 successor, TX-2, is storied for its role in pioneering human-computer interaction and AI. With TX-GAIN, the LLSC continues this legacy. More