Research

Subterms

    Latest story

    138 Shares119 Views

    Method prevents an AI model from being overconfident about wrong answers

    by

    People use large language models for a huge array of tasks, from translating an article to identifying financial fraud. However, despite the incredible capabilities and versatility of these models, they sometimes generate inaccurate responses.On top of that problem, the models can be overconfident about wrong answers or underconfident about correct ones, making it tough for a user to know when a model can be trusted.Researchers typically calibrate a machine-learning model to ensure its level of confidence lines up with its accuracy. A well-calibrated model should have less confidence about an incorrect prediction, and vice-versa. But because large language models (LLMs) can be applied to a seemingly endless collection of diverse tasks, traditional calibration methods are ineffective.Now, researchers from MIT and the MIT-IBM Watson AI Lab have introduced a calibration method tailored to large language models. Their method, called Thermometer, involves building a smaller, auxiliary model that runs on top of a large language model to calibrate it.Thermometer is more efficient than other approaches — requiring less power-hungry computation — while preserving the accuracy of the model and enabling it to produce better-calibrated responses on tasks it has not seen before.By enabling efficient calibration of an LLM for a variety of tasks, Thermometer could help users pinpoint situations where a model is overconfident about false predictions, ultimately preventing them from deploying that model in a situation where it may fail.“With Thermometer, we want to provide the user with a clear signal to tell them whether a model’s response is accurate or inaccurate, in a way that reflects the model’s uncertainty, so they know if that model is reliable,” says Maohao Shen, an electrical engineering and computer science (EECS) graduate student and lead author of a paper on Thermometer.Shen is joined on the paper by Gregory Wornell, the Sumitomo Professor of Engineering who leads the Signals, Information, and Algorithms Laboratory in the Research Laboratory for Electronics, and is a member of the MIT-IBM Watson AI Lab; senior author Soumya Ghosh, a research staff member in the MIT-IBM Watson AI Lab; as well as others at MIT and the MIT-IBM Watson AI Lab. The research was recently presented at the International Conference on Machine Learning.Universal calibrationSince traditional machine-learning models are typically designed to perform a single task, calibrating them usually involves one task-specific method. On the other hand, since LLMs have the flexibility to perform many tasks, using a traditional method to calibrate that model for one task might hurt its performance on another task.Calibrating an LLM often involves sampling from the model multiple times to obtain different predictions and then aggregating these predictions to obtain better-calibrated confidence. However, because these models have billions of parameters, the computational costs of such approaches rapidly add up.“In a sense, large language models are universal because they can handle various tasks. So, we need a universal calibration method that can also handle many different tasks,” says Shen.With Thermometer, the researchers developed a versatile technique that leverages a classical calibration method called temperature scaling to efficiently calibrate an LLM for a new task.In this context, a “temperature” is a scaling parameter used to adjust a model’s confidence to be aligned with its prediction accuracy. Traditionally, one determines the right temperature using a labeled validation dataset of task-specific examples.Since LLMs are often applied to new tasks, labeled datasets can be nearly impossible to acquire. For instance, a user who wants to deploy an LLM to answer customer questions about a new product likely does not have a dataset containing such questions and answers.Instead of using a labeled dataset, the researchers train an auxiliary model that runs on top of an LLM to automatically predict the temperature needed to calibrate it for this new task.They use labeled datasets of a few representative tasks to train the Thermometer model, but then once it has been trained, it can generalize to new tasks in a similar category without the need for additional labeled data.A Thermometer model trained on a collection of multiple-choice question datasets, perhaps including one with algebra questions and one with medical questions, could be used to calibrate an LLM that will answer questions about geometry or biology, for instance.“The aspirational goal is for it to work on any task, but we are not quite there yet,” Ghosh says.   The Thermometer model only needs to access a small part of the LLM’s inner workings to predict the right temperature that will calibrate its prediction for data points of a specific task. An efficient approachImportantly, the technique does not require multiple training runs and only slightly slows the LLM. Plus, since temperature scaling does not alter a model’s predictions, Thermometer preserves its accuracy.When they compared Thermometer to several baselines on multiple tasks, it consistently produced better-calibrated uncertainty measures while requiring much less computation.“As long as we train a Thermometer model on a sufficiently large number of tasks, it should be able to generalize well across any new task, just like a large language model, it is also a universal model,” Shen adds.The researchers also found that if they train a Thermometer model for a smaller LLM, it can be directly applied to calibrate a larger LLM within the same family.In the future, they want to adapt Thermometer for more complex text-generation tasks and apply the technique to even larger LLMs. The researchers also hope to quantify the diversity and number of labeled datasets one would need to train a Thermometer model so it can generalize to a new task.This research was funded, in part, by the MIT-IBM Watson AI Lab. More

    More stories

    • 138 Shares189 Views

      in Data Management & Statistics

      Study: When allocating scarce resources with AI, randomization can improve fairness

      by

      Organizations are increasingly utilizing machine-learning models to allocate scarce resources or opportunities. For instance, such models can help companies screen resumes to choose job interview candidates or aid hospitals in ranking kidney transplant patients based on their likelihood of survival.When deploying a model, users typically strive to ensure its predictions are fair by reducing bias. This often involves techniques like adjusting the features a model uses to make decisions or calibrating the scores it generates.However, researchers from MIT and Northeastern University argue that these fairness methods are not sufficient to address structural injustices and inherent uncertainties. In a new paper, they show how randomizing a model’s decisions in a structured way can improve fairness in certain situations.For example, if multiple companies use the same machine-learning model to rank job interview candidates deterministically — without any randomization — then one deserving individual could be the bottom-ranked candidate for every job, perhaps due to how the model weighs answers provided in an online form. Introducing randomization into a model’s decisions could prevent one worthy person or group from always being denied a scarce resource, like a job interview.Through their analysis, the researchers found that randomization can be especially beneficial when a model’s decisions involve uncertainty or when the same group consistently receives negative decisions.They present a framework one could use to introduce a specific amount of randomization into a model’s decisions by allocating resources through a weighted lottery. This method, which an individual can tailor to fit their situation, can improve fairness without hurting the efficiency or accuracy of a model.“Even if you could make fair predictions, should you be deciding these social allocations of scarce resources or opportunities strictly off scores or rankings? As things scale, and we see more and more opportunities being decided by these algorithms, the inherent uncertainties in these scores can be amplified. We show that fairness may require some sort of randomization,” says Shomik Jain, a graduate student in the Institute for Data, Systems, and Society (IDSS) and lead author of the paper.Jain is joined on the paper by Kathleen Creel, assistant professor of philosophy and computer science at Northeastern University; and senior author Ashia Wilson, the Lister Brothers Career Development Professor in the Department of Electrical Engineering and Computer Science and a principal investigator in the Laboratory for Information and Decision Systems (LIDS). The research will be presented at the International Conference on Machine Learning.Considering claimsThis work builds off a previous paper in which the researchers explored harms that can occur when one uses deterministic systems at scale. They found that using a machine-learning model to deterministically allocate resources can amplify inequalities that exist in training data, which can reinforce bias and systemic inequality. “Randomization is a very useful concept in statistics, and to our delight, satisfies the fairness demands coming from both a systemic and individual point of view,” Wilson says.In this paper, they explored the question of when randomization can improve fairness. They framed their analysis around the ideas of philosopher John Broome, who wrote about the value of using lotteries to award scarce resources in a way that honors all claims of individuals.A person’s claim to a scarce resource, like a kidney transplant, can stem from merit, deservingness, or need. For instance, everyone has a right to life, and their claims on a kidney transplant may stem from that right, Wilson explains.“When you acknowledge that people have different claims to these scarce resources, fairness is going to require that we respect all claims of individuals. If we always give someone with a stronger claim the resource, is that fair?” Jain says.That sort of deterministic allocation could cause systemic exclusion or exacerbate patterned inequality, which occurs when receiving one allocation increases an individual’s likelihood of receiving future allocations. In addition, machine-learning models can make mistakes, and a deterministic approach could cause the same mistake to be repeated.Randomization can overcome these problems, but that doesn’t mean all decisions a model makes should be randomized equally.Structured randomizationThe researchers use a weighted lottery to adjust the level of randomization based on the amount of uncertainty involved in the model’s decision-making. A decision that is less certain should incorporate more randomization.“In kidney allocation, usually the planning is around projected lifespan, and that is deeply uncertain. If two patients are only five years apart, it becomes a lot harder to measure. We want to leverage that level of uncertainty to tailor the randomization,” Wilson says.The researchers used statistical uncertainty quantification methods to determine how much randomization is needed in different situations. They show that calibrated randomization can lead to fairer outcomes for individuals without significantly affecting the utility, or effectiveness, of the model.“There is a balance to be had between overall utility and respecting the rights of the individuals who are receiving a scarce resource, but oftentimes the tradeoff is relatively small,” says Wilson.However, the researchers emphasize there are situations where randomizing decisions would not improve fairness and could harm individuals, such as in criminal justice contexts.But there could be other areas where randomization can improve fairness, such as college admissions, and the researchers plan to study other use cases in future work. They also want to explore how randomization can affect other factors, such as competition or prices, and how it could be used to improve the robustness of machine-learning models.“We are hoping our paper is a first move toward illustrating that there might be a benefit to randomization. We are offering randomization as a tool. How much you are going to want to do it is going to be up to all the stakeholders in the allocation to decide. And, of course, how they decide is another research question all together,” says Wilson. More

    • 163 Shares169 Views

      in Data Management & Statistics

      AI model identifies certain breast tumor stages likely to progress to invasive cancer

      by

      Ductal carcinoma in situ (DCIS) is a type of preinvasive tumor that sometimes progresses to a highly deadly form of breast cancer. It accounts for about 25 percent of all breast cancer diagnoses.Because it is difficult for clinicians to determine the type and stage of DCIS, patients with DCIS are often overtreated. To address this, an interdisciplinary team of researchers from MIT and ETH Zurich developed an AI model that can identify the different stages of DCIS from a cheap and easy-to-obtain breast tissue image. Their model shows that both the state and arrangement of cells in a tissue sample are important for determining the stage of DCIS.Because such tissue images are so easy to obtain, the researchers were able to build one of the largest datasets of its kind, which they used to train and test their model. When they compared its predictions to conclusions of a pathologist, they found clear agreement in many instances.In the future, the model could be used as a tool to help clinicians streamline the diagnosis of simpler cases without the need for labor-intensive tests, giving them more time to evaluate cases where it is less clear if DCIS will become invasive.“We took the first step in understanding that we should be looking at the spatial organization of cells when diagnosing DCIS, and now we have developed a technique that is scalable. From here, we really need a prospective study. Working with a hospital and getting this all the way to the clinic will be an important step forward,” says Caroline Uhler, a professor in the Department of Electrical Engineering and Computer Science (EECS) and the Institute for Data, Systems, and Society (IDSS), who is also director of the Eric and Wendy Schmidt Center at the Broad Institute of MIT and Harvard and a researcher at MIT’s Laboratory for Information and Decision Systems (LIDS).Uhler, co-corresponding author of a paper on this research, is joined by lead author Xinyi Zhang, a graduate student in EECS and the Eric and Wendy Schmidt Center; co-corresponding author GV Shivashankar, professor of mechogenomics at ETH Zurich jointly with the Paul Scherrer Institute; and others at MIT, ETH Zurich, and the University of Palermo in Italy. The open-access research was published July 20 in Nature Communications.Combining imaging with AI   Between 30 and 50 percent of patients with DCIS develop a highly invasive stage of cancer, but researchers don’t know the biomarkers that could tell a clinician which tumors will progress.Researchers can use techniques like multiplexed staining or single-cell RNA sequencing to determine the stage of DCIS in tissue samples. However, these tests are too expensive to be performed widely, Shivashankar explains.In previous work, these researchers showed that a cheap imagining technique known as chromatin staining could be as informative as the much costlier single-cell RNA sequencing.For this research, they hypothesized that combining this single stain with a carefully designed machine-learning model could provide the same information about cancer stage as costlier techniques.First, they created a dataset containing 560 tissue sample images from 122 patients at three different stages of disease. They used this dataset to train an AI model that learns a representation of the state of each cell in a tissue sample image, which it uses to infer the stage of a patient’s cancer.However, not every cell is indicative of cancer, so the researchers had to aggregate them in a meaningful way.They designed the model to create clusters of cells in similar states, identifying eight states that are important markers of DCIS. Some cell states are more indicative of invasive cancer than others. The model determines the proportion of cells in each state in a tissue sample.Organization matters“But in cancer, the organization of cells also changes. We found that just having the proportions of cells in every state is not enough. You also need to understand how the cells are organized,” says Shivashankar.With this insight, they designed the model to consider proportion and arrangement of cell states, which significantly boosted its accuracy.“The interesting thing for us was seeing how much spatial organization matters. Previous studies had shown that cells which are close to the breast duct are important. But it is also important to consider which cells are close to which other cells,” says Zhang.When they compared the results of their model with samples evaluated by a pathologist, it had clear agreement in many instances. In cases that were not as clear-cut, the model could provide information about features in a tissue sample, like the organization of cells, that a pathologist could use in decision-making.This versatile model could also be adapted for use in other types of cancer, or even neurodegenerative conditions, which is one area the researchers are also currently exploring.“We have shown that, with the right AI techniques, this simple stain can be very powerful. There is still much more research to do, but we need to take the organization of cells into account in more of our studies,” Uhler says.This research was funded, in part, by the Eric and Wendy Schmidt Center at the Broad Institute, ETH Zurich, the Paul Scherrer Institute, the Swiss National Science Foundation, the U.S. National Institutes of Health, the U.S. Office of Naval Research, the MIT Jameel Clinic for Machine Learning and Health, the MIT-IBM Watson AI Lab, and a Simons Investigator Award. More

    • 138 Shares109 Views

      in Data Management & Statistics

      How to assess a general-purpose AI model’s reliability before it’s deployed

      by

      Foundation models are massive deep-learning models that have been pretrained on an enormous amount of general-purpose, unlabeled data. They can be applied to a variety of tasks, like generating images or answering customer questions.But these models, which serve as the backbone for powerful artificial intelligence tools like ChatGPT and DALL-E, can offer up incorrect or misleading information. In a safety-critical situation, such as a pedestrian approaching a self-driving car, these mistakes could have serious consequences.To help prevent such mistakes, researchers from MIT and the MIT-IBM Watson AI Lab developed a technique to estimate the reliability of foundation models before they are deployed to a specific task.They do this by considering a set of foundation models that are slightly different from one another. Then they use their algorithm to assess the consistency of the representations each model learns about the same test data point. If the representations are consistent, it means the model is reliable.When they compared their technique to state-of-the-art baseline methods, it was better at capturing the reliability of foundation models on a variety of downstream classification tasks.Someone could use this technique to decide if a model should be applied in a certain setting, without the need to test it on a real-world dataset. This could be especially useful when datasets may not be accessible due to privacy concerns, like in health care settings. In addition, the technique could be used to rank models based on reliability scores, enabling a user to select the best one for their task.“All models can be wrong, but models that know when they are wrong are more useful. The problem of quantifying uncertainty or reliability is more challenging for these foundation models because their abstract representations are difficult to compare. Our method allows one to quantify how reliable a representation model is for any given input data,” says senior author Navid Azizan, the Esther and Harold E. Edgerton Assistant Professor in the MIT Department of Mechanical Engineering and the Institute for Data, Systems, and Society (IDSS), and a member of the Laboratory for Information and Decision Systems (LIDS).He is joined on a paper about the work by lead author Young-Jin Park, a LIDS graduate student; Hao Wang, a research scientist at the MIT-IBM Watson AI Lab; and Shervin Ardeshir, a senior research scientist at Netflix. The paper will be presented at the Conference on Uncertainty in Artificial Intelligence.Measuring consensusTraditional machine-learning models are trained to perform a specific task. These models typically make a concrete prediction based on an input. For instance, the model might tell you whether a certain image contains a cat or a dog. In this case, assessing reliability could be a matter of looking at the final prediction to see if the model is right.But foundation models are different. The model is pretrained using general data, in a setting where its creators don’t know all downstream tasks it will be applied to. Users adapt it to their specific tasks after it has already been trained.Unlike traditional machine-learning models, foundation models don’t give concrete outputs like “cat” or “dog” labels. Instead, they generate an abstract representation based on an input data point.To assess the reliability of a foundation model, the researchers used an ensemble approach by training several models which share many properties but are slightly different from one another.“Our idea is like measuring the consensus. If all those foundation models are giving consistent representations for any data in our dataset, then we can say this model is reliable,” Park says.But they ran into a problem: How could they compare abstract representations?“These models just output a vector, comprised of some numbers, so we can’t compare them easily,” he adds.They solved this problem using an idea called neighborhood consistency.For their approach, the researchers prepare a set of reliable reference points to test on the ensemble of models. Then, for each model, they investigate the reference points located near that model’s representation of the test point.By looking at the consistency of neighboring points, they can estimate the reliability of the models.Aligning the representationsFoundation models map data points to what is known as a representation space. One way to think about this space is as a sphere. Each model maps similar data points to the same part of its sphere, so images of cats go in one place and images of dogs go in another.But each model would map animals differently in its own sphere, so while cats may be grouped near the South Pole of one sphere, another model could map cats somewhere in the Northern Hemisphere.The researchers use the neighboring points like anchors to align those spheres so they can make the representations comparable. If a data point’s neighbors are consistent across multiple representations, then one should be confident about the reliability of the model’s output for that point.When they tested this approach on a wide range of classification tasks, they found that it was much more consistent than baselines. Plus, it wasn’t tripped up by challenging test points that caused other methods to fail.Moreover, their approach can be used to assess reliability for any input data, so one could evaluate how well a model works for a particular type of individual, such as a patient with certain characteristics.“Even if the models all have average performance overall, from an individual point of view, you’d prefer the one that works best for that individual,” Wang says.However, one limitation comes from the fact that they must train an ensemble of foundation models, which is computationally expensive. In the future, they plan to find more efficient ways to build multiple models, perhaps by using small perturbations of a single model.“With the current trend of using foundational models for their embeddings to support various downstream tasks — from fine-tuning to retrieval augmented generation — the topic of quantifying uncertainty at the representation level is increasingly important, but challenging, as embeddings on their own have no grounding. What matters instead is how embeddings of different inputs are related to one another, an idea that this work neatly captures through the proposed neighborhood consistency score,” says Marco Pavone, an associate professor in the Department of Aeronautics and Astronautics at Stanford University, who was not involved with this work. “This is a promising step towards high quality uncertainty quantifications for embedding models, and I’m excited to see future extensions which can operate without requiring model-ensembling to really enable this approach to scale to foundation-size models.”This work is funded, in part, by the MIT-IBM Watson AI Lab, MathWorks, and Amazon. More

    • 75 Shares169 Views

      in Data Management & Statistics

      When to trust an AI model

      by

      Because machine-learning models can give false predictions, researchers often equip them with the ability to tell a user how confident they are about a certain decision. This is especially important in high-stake settings, such as when models are used to help identify disease in medical images or filter job applications.But a model’s uncertainty quantifications are only useful if they are accurate. If a model says it is 49 percent confident that a medical image shows a pleural effusion, then 49 percent of the time, the model should be right.MIT researchers have introduced a new approach that can improve uncertainty estimates in machine-learning models. Their method not only generates more accurate uncertainty estimates than other techniques, but does so more efficiently.In addition, because the technique is scalable, it can be applied to huge deep-learning models that are increasingly being deployed in health care and other safety-critical situations.This technique could give end users, many of whom lack machine-learning expertise, better information they can use to determine whether to trust a model’s predictions or if the model should be deployed for a particular task.“It is easy to see these models perform really well in scenarios where they are very good, and then assume they will be just as good in other scenarios. This makes it especially important to push this kind of work that seeks to better calibrate the uncertainty of these models to make sure they align with human notions of uncertainty,” says lead author Nathan Ng, a graduate student at the University of Toronto who is a visiting student at MIT.Ng wrote the paper with Roger Grosse, an assistant professor of computer science at the University of Toronto; and senior author Marzyeh Ghassemi, an associate professor in the Department of Electrical Engineering and Computer Science and a member of the Institute of Medical Engineering Sciences and the Laboratory for Information and Decision Systems. The research will be presented at the International Conference on Machine Learning.Quantifying uncertaintyUncertainty quantification methods often require complex statistical calculations that don’t scale well to machine-learning models with millions of parameters. These methods also require users to make assumptions about the model and data used to train it.The MIT researchers took a different approach. They use what is known as the minimum description length principle (MDL), which does not require the assumptions that can hamper the accuracy of other methods. MDL is used to better quantify and calibrate uncertainty for test points the model has been asked to label.The technique the researchers developed, known as IF-COMP, makes MDL fast enough to use with the kinds of large deep-learning models deployed in many real-world settings.MDL involves considering all possible labels a model could give a test point. If there are many alternative labels for this point that fit well, its confidence in the label it chose should decrease accordingly.“One way to understand how confident a model is would be to tell it some counterfactual information and see how likely it is to believe you,” Ng says.For example, consider a model that says a medical image shows a pleural effusion. If the researchers tell the model this image shows an edema, and it is willing to update its belief, then the model should be less confident in its original decision.With MDL, if a model is confident when it labels a datapoint, it should use a very short code to describe that point. If it is uncertain about its decision because the point could have many other labels, it uses a longer code to capture these possibilities.The amount of code used to label a datapoint is known as stochastic data complexity. If the researchers ask the model how willing it is to update its belief about a datapoint given contrary evidence, the stochastic data complexity should decrease if the model is confident.But testing each datapoint using MDL would require an enormous amount of computation.Speeding up the processWith IF-COMP, the researchers developed an approximation technique that can accurately estimate stochastic data complexity using a special function, known as an influence function. They also employed a statistical technique called temperature-scaling, which improves the calibration of the model’s outputs. This combination of influence functions and temperature-scaling enables high-quality approximations of the stochastic data complexity.In the end, IF-COMP can efficiently produce well-calibrated uncertainty quantifications that reflect a model’s true confidence. The technique can also determine whether the model has mislabeled certain data points or reveal which data points are outliers.The researchers tested their system on these three tasks and found that it was faster and more accurate than other methods.“It is really important to have some certainty that a model is well-calibrated, and there is a growing need to detect when a specific prediction doesn’t look quite right. Auditing tools are becoming more necessary in machine-learning problems as we use large amounts of unexamined data to make models that will be applied to human-facing problems,” Ghassemi says.IF-COMP is model-agnostic, so it can provide accurate uncertainty quantifications for many types of machine-learning models. This could enable it to be deployed in a wider range of real-world settings, ultimately helping more practitioners make better decisions.“People need to understand that these systems are very fallible and can make things up as they go. A model may look like it is highly confident, but there are a ton of different things it is willing to believe given evidence to the contrary,” Ng says.In the future, the researchers are interested in applying their approach to large language models and studying other potential use cases for the minimum description length principle.  More

    • 150 Shares169 Views

      in Data Management & Statistics

      Study finds health risks in switching ships from diesel to ammonia fuel

      by

      As container ships the size of city blocks cross the oceans to deliver cargo, their huge diesel engines emit large quantities of air pollutants that drive climate change and have human health impacts. It has been estimated that maritime shipping accounts for almost 3 percent of global carbon dioxide emissions and the industry’s negative impacts on air quality cause about 100,000 premature deaths each year.Decarbonizing shipping to reduce these detrimental effects is a goal of the International Maritime Organization, a U.N. agency that regulates maritime transport. One potential solution is switching the global fleet from fossil fuels to sustainable fuels such as ammonia, which could be nearly carbon-free when considering its production and use.But in a new study, an interdisciplinary team of researchers from MIT and elsewhere caution that burning ammonia for maritime fuel could worsen air quality further and lead to devastating public health impacts, unless it is adopted alongside strengthened emissions regulations.Ammonia combustion generates nitrous oxide (N2O), a greenhouse gas that is about 300 times more potent than carbon dioxide. It also emits nitrogen in the form of nitrogen oxides (NO and NO2, referred to as NOx), and unburnt ammonia may slip out, which eventually forms fine particulate matter in the atmosphere. These tiny particles can be inhaled deep into the lungs, causing health problems like heart attacks, strokes, and asthma.The new study indicates that, under current legislation, switching the global fleet to ammonia fuel could cause up to about 600,000 additional premature deaths each year. However, with stronger regulations and cleaner engine technology, the switch could lead to about 66,000 fewer premature deaths than currently caused by maritime shipping emissions, with far less impact on global warming.“Not all climate solutions are created equal. There is almost always some price to pay. We have to take a more holistic approach and consider all the costs and benefits of different climate solutions, rather than just their potential to decarbonize,” says Anthony Wong, a postdoc in the MIT Center for Global Change Science and lead author of the study.His co-authors include Noelle Selin, an MIT professor in the Institute for Data, Systems, and Society and the Department of Earth, Atmospheric and Planetary Sciences (EAPS); Sebastian Eastham, a former principal research scientist who is now a senior lecturer at Imperial College London; Christine Mounaïm-Rouselle, a professor at the University of Orléans in France; Yiqi Zhang, a researcher at the Hong Kong University of Science and Technology; and Florian Allroggen, a research scientist in the MIT Department of Aeronautics and Astronautics. The research appears this week in Environmental Research Letters.Greener, cleaner ammoniaTraditionally, ammonia is made by stripping hydrogen from natural gas and then combining it with nitrogen at extremely high temperatures. This process is often associated with a large carbon footprint. The maritime shipping industry is betting on the development of “green ammonia,” which is produced by using renewable energy to make hydrogen via electrolysis and to generate heat.“In theory, if you are burning green ammonia in a ship engine, the carbon emissions are almost zero,” Wong says.But even the greenest ammonia generates nitrous oxide (N2O), nitrogen oxides (NOx) when combusted, and some of the ammonia may slip out, unburnt. This nitrous oxide would escape into the atmosphere, where the greenhouse gas would remain for more than 100 years. At the same time, the nitrogen emitted as NOx and ammonia would fall to Earth, damaging fragile ecosystems. As these emissions are digested by bacteria, additional N2O  is produced.NOx and ammonia also mix with gases in the air to form fine particulate matter. A primary contributor to air pollution, fine particulate matter kills an estimated 4 million people each year.“Saying that ammonia is a ‘clean’ fuel is a bit of an overstretch. Just because it is carbon-free doesn’t necessarily mean it is clean and good for public health,” Wong says.A multifaceted modelThe researchers wanted to paint the whole picture, capturing the environmental and public health impacts of switching the global fleet to ammonia fuel. To do so, they designed scenarios to measure how pollutant impacts change under certain technology and policy assumptions.From a technological point of view, they considered two ship engines. The first burns pure ammonia, which generates higher levels of unburnt ammonia but emits fewer nitrogen oxides. The second engine technology involves mixing ammonia with hydrogen to improve combustion and optimize the performance of a catalytic converter, which controls both nitrogen oxides and unburnt ammonia pollution.They also considered three policy scenarios: current regulations, which only limit NOx emissions in some parts of the world; a scenario that adds ammonia emission limits over North America and Western Europe; and a scenario that adds global limits on ammonia and NOx emissions.The researchers used a ship track model to calculate how pollutant emissions change under each scenario and then fed the results into an air quality model. The air quality model calculates the impact of ship emissions on particulate matter and ozone pollution. Finally, they estimated the effects on global public health.One of the biggest challenges came from a lack of real-world data, since no ammonia-powered ships are yet sailing the seas. Instead, the researchers relied on experimental ammonia combustion data from collaborators to build their model.“We had to come up with some clever ways to make that data useful and informative to both the technology and regulatory situations,” he says.A range of outcomesIn the end, they found that with no new regulations and ship engines that burn pure ammonia, switching the entire fleet would cause 681,000 additional premature deaths each year.“While a scenario with no new regulations is not very realistic, it serves as a good warning of how dangerous ammonia emissions could be. And unlike NOx, ammonia emissions from shipping are currently unregulated,” Wong says.However, even without new regulations, using cleaner engine technology would cut the number of premature deaths down to about 80,000, which is about 20,000 fewer than are currently attributed to maritime shipping emissions. With stronger global regulations and cleaner engine technology, the number of people killed by air pollution from shipping could be reduced by about 66,000.“The results of this study show the importance of developing policies alongside new technologies,” Selin says. “There is a potential for ammonia in shipping to be beneficial for both climate and air quality, but that requires that regulations be designed to address the entire range of potential impacts, including both climate and air quality.”Ammonia’s air quality impacts would not be felt uniformly across the globe, and addressing them fully would require coordinated strategies across very different contexts. Most premature deaths would occur in East Asia, since air quality regulations are less stringent in this region. Higher levels of existing air pollution cause the formation of more particulate matter from ammonia emissions. In addition, shipping volume over East Asia is far greater than elsewhere on Earth, compounding these negative effects.In the future, the researchers want to continue refining their analysis. They hope to use these findings as a starting point to urge the marine industry to share engine data they can use to better evaluate air quality and climate impacts. They also hope to inform policymakers about the importance and urgency of updating shipping emission regulations.This research was funded by the MIT Climate and Sustainability Consortium. More

    • 88 Shares149 Views

      in Data Management & Statistics

      MIT researchers introduce generative AI for databases

      by

      A new tool makes it easier for database users to perform complicated statistical analyses of tabular data without the need to know what is going on behind the scenes.GenSQL, a generative AI system for databases, could help users make predictions, detect anomalies, guess missing values, fix errors, or generate synthetic data with just a few keystrokes.For instance, if the system were used to analyze medical data from a patient who has always had high blood pressure, it could catch a blood pressure reading that is low for that particular patient but would otherwise be in the normal range.GenSQL automatically integrates a tabular dataset and a generative probabilistic AI model, which can account for uncertainty and adjust their decision-making based on new data.Moreover, GenSQL can be used to produce and analyze synthetic data that mimic the real data in a database. This could be especially useful in situations where sensitive data cannot be shared, such as patient health records, or when real data are sparse.This new tool is built on top of SQL, a programming language for database creation and manipulation that was introduced in the late 1970s and is used by millions of developers worldwide.“Historically, SQL taught the business world what a computer could do. They didn’t have to write custom programs, they just had to ask questions of a database in high-level language. We think that, when we move from just querying data to asking questions of models and data, we are going to need an analogous language that teaches people the coherent questions you can ask a computer that has a probabilistic model of the data,” says Vikash Mansinghka ’05, MEng ’09, PhD ’09, senior author of a paper introducing GenSQL and a principal research scientist and leader of the Probabilistic Computing Project in the MIT Department of Brain and Cognitive Sciences.When the researchers compared GenSQL to popular, AI-based approaches for data analysis, they found that it was not only faster but also produced more accurate results. Importantly, the probabilistic models used by GenSQL are explainable, so users can read and edit them.“Looking at the data and trying to find some meaningful patterns by just using some simple statistical rules might miss important interactions. You really want to capture the correlations and the dependencies of the variables, which can be quite complicated, in a model. With GenSQL, we want to enable a large set of users to query their data and their model without having to know all the details,” adds lead author Mathieu Huot, a research scientist in the Department of Brain and Cognitive Sciences and member of the Probabilistic Computing Project.They are joined on the paper by Matin Ghavami and Alexander Lew, MIT graduate students; Cameron Freer, a research scientist; Ulrich Schaechtel and Zane Shelby of Digital Garage; Martin Rinard, an MIT professor in the Department of Electrical Engineering and Computer Science and member of the Computer Science and Artificial Intelligence Laboratory (CSAIL); and Feras Saad ’15, MEng ’16, PhD ’22, an assistant professor at Carnegie Mellon University. The research was recently presented at the ACM Conference on Programming Language Design and Implementation.Combining models and databasesSQL, which stands for structured query language, is a programming language for storing and manipulating information in a database. In SQL, people can ask questions about data using keywords, such as by summing, filtering, or grouping database records.However, querying a model can provide deeper insights, since models can capture what data imply for an individual. For instance, a female developer who wonders if she is underpaid is likely more interested in what salary data mean for her individually than in trends from database records.The researchers noticed that SQL didn’t provide an effective way to incorporate probabilistic AI models, but at the same time, approaches that use probabilistic models to make inferences didn’t support complex database queries.They built GenSQL to fill this gap, enabling someone to query both a dataset and a probabilistic model using a straightforward yet powerful formal programming language.A GenSQL user uploads their data and probabilistic model, which the system automatically integrates. Then, she can run queries on data that also get input from the probabilistic model running behind the scenes. This not only enables more complex queries but can also provide more accurate answers.For instance, a query in GenSQL might be something like, “How likely is it that a developer from Seattle knows the programming language Rust?” Just looking at a correlation between columns in a database might miss subtle dependencies. Incorporating a probabilistic model can capture more complex interactions.   Plus, the probabilistic models GenSQL utilizes are auditable, so people can see which data the model uses for decision-making. In addition, these models provide measures of calibrated uncertainty along with each answer.For instance, with this calibrated uncertainty, if one queries the model for predicted outcomes of different cancer treatments for a patient from a minority group that is underrepresented in the dataset, GenSQL would tell the user that it is uncertain, and how uncertain it is, rather than overconfidently advocating for the wrong treatment.Faster and more accurate resultsTo evaluate GenSQL, the researchers compared their system to popular baseline methods that use neural networks. GenSQL was between 1.7 and 6.8 times faster than these approaches, executing most queries in a few milliseconds while providing more accurate results.They also applied GenSQL in two case studies: one in which the system identified mislabeled clinical trial data and the other in which it generated accurate synthetic data that captured complex relationships in genomics.Next, the researchers want to apply GenSQL more broadly to conduct largescale modeling of human populations. With GenSQL, they can generate synthetic data to draw inferences about things like health and salary while controlling what information is used in the analysis.They also want to make GenSQL easier to use and more powerful by adding new optimizations and automation to the system. In the long run, the researchers want to enable users to make natural language queries in GenSQL. Their goal is to eventually develop a ChatGPT-like AI expert one could talk to about any database, which grounds its answers using GenSQL queries.   This research is funded, in part, by the Defense Advanced Research Projects Agency (DARPA), Google, and the Siegel Family Foundation. More

    • 150 Shares199 Views

      in Data Management & Statistics

      Fotini Christia named director of the Institute for Data, Systems, and Society

      by

      Fotini Christia, the Ford International Professor of Social Sciences in the Department of Political Science, has been named the new director of the Institute for Data, Systems, and Society (IDSS), effective July 1.“Fotini is well-positioned to guide IDSS into the next chapter. With her tenure as the director of the Sociotechnical Systems Research Center and as an associate director of IDSS since 2020, she has actively forged connections between the social sciences, data science, and computation,” says Daniel Huttenlocher, dean of the MIT Schwarzman College of Computing and the Henry Ellis Warren Professor of Electrical Engineering and Computer Science. “I eagerly anticipate the ways in which she will advance and champion IDSS in alignment with the spirit and mission of the Schwarzman College of Computing.”“Fotini’s profound expertise as a social scientist and her adept use of data science, computational tools, and novel methodologies to grasp the dynamics of societal evolution across diverse fields, makes her a natural fit to lead IDSS,” says Asu Ozdaglar, deputy dean of the MIT Schwarzman College of Computing and head of the Department of Electrical Engineering and Computer Science.Christia’s research has focused on issues of conflict and cooperation in the Muslim world, for which she has conducted fieldwork in Afghanistan, Bosnia, Iraq, the Palestinian Territories, and Yemen, among others. More recently, her research has been directed at examining how to effectively integrate artificial intelligence tools in public policy.She was appointed the director of the Sociotechnical Systems Research Center (SSRC) and an associate director of IDSS in October 2020. SSRC, an interdisciplinary center housed within IDSS in the MIT Schwarzman College of Computing, focuses on the study of high-impact, complex societal challenges that shape our world.As part of IDSS, she is co-organizer of a cross-disciplinary research effort, the Initiative on Combatting Systemic Racism. Bringing together faculty and researchers from all of MIT’s five schools and the college, the initiative builds on extensive social science literature on systemic racism and uses big data to develop and harness computational tools that can help effect structural and normative change toward racial equity across housing, health care, policing, and social media. Christia is also chair of IDSS’s doctoral program in Social and Engineering Systems.Christia is the author of “Alliance Formation in Civil War” (Cambridge University Press, 2012), which was awarded the Luebbert Award for Best Book in Comparative Politics, the Lepgold Prize for Best Book in International Relations, and a Distinguished Book Award from the International Studies Association. She is co-editor with Graeme Blair (University of California, Los Angeles) and Jeremy Weinstein (incoming dean at Harvard Kennedy School) of “Crime, Insecurity, and Community Policing: Experiments on Building Trust,” forthcoming in August 2024 with Cambridge University Press.Her research has also appeared in Science, Nature Human Behavior, Review of Economic Studies, American Economic Journal: Applied Economics, NeurIPs, Communications Medicine, IEEE Transactions on Network Science and Engineering, American Political Science Review, and Annual Review of Political Science, among other journals. Her opinion pieces have been published in Foreign Affairs, The New York Times, The Washington Post, and The Boston Globe, among other outlets.A native of Greece, where she grew up in the port city of Salonika, Christia moved to the United States to attend college at Columbia University. She graduated magna cum laude in 2001 with a joint BA in economics–operations research and an MA in international affairs. She joined the MIT faculty in 2008 after receiving her PhD in public policy from Harvard University.Christia succeeds Noelle Selin, a professor in IDSS and the Department of Earth, Atmospheric, and Planetary Sciences. Selin has led IDSS as interim director for the 2023-24 academic year since July 2023, following Professor Martin Wainwright.“I am incredibly grateful to Noelle for serving as interim director this year. Her contributions in this role, as well as her time leading the Technology and Policy Program, have been invaluable. I’m delighted she will remain part of the IDSS community as a faculty member,” says Huttenlocher. More

    • 88 Shares159 Views

      in Data Management & Statistics

      Scientists preserve DNA in an amber-like polymer

      by

      In the movie “Jurassic Park,” scientists extracted DNA that had been preserved in amber for millions of years, and used it to create a population of long-extinct dinosaurs.Inspired partly by that film, MIT researchers have developed a glassy, amber-like polymer that can be used for long-term storage of DNA, whether entire human genomes or digital files such as photos.Most current methods for storing DNA require freezing temperatures, so they consume a great deal of energy and are not feasible in many parts of the world. In contrast, the new amber-like polymer can store DNA at room temperature while protecting the molecules from damage caused by heat or water.The researchers showed that they could use this polymer to store DNA sequences encoding the theme music from Jurassic Park, as well as an entire human genome. They also demonstrated that the DNA can be easily removed from the polymer without damaging it.“Freezing DNA is the number one way to preserve it, but it’s very expensive, and it’s not scalable,” says James Banal, a former MIT postdoc. “I think our new preservation method is going to be a technology that may drive the future of storing digital information on DNA.”Banal and Jeremiah Johnson, the A. Thomas Geurtin Professor of Chemistry at MIT, are the senior authors of the study, published yesterday in the Journal of the American Chemical Society. Former MIT postdoc Elizabeth Prince and MIT postdoc Ho Fung Cheng are the lead authors of the paper.Capturing DNADNA, a very stable molecule, is well-suited for storing massive amounts of information, including digital data. Digital storage systems encode text, photos, and other kind of information as a series of 0s and 1s. This same information can be encoded in DNA using the four nucleotides that make up the genetic code: A, T, G, and C. For example, G and C could be used to represent 0 while A and T represent 1.DNA offers a way to store this digital information at very high density: In theory, a coffee mug full of DNA could store all of the world’s data. DNA is also very stable and relatively easy to synthesize and sequence.In 2021, Banal and his postdoc advisor, Mark Bathe, an MIT professor of biological engineering, developed a way to store DNA in particles of silica, which could be labeled with tags that revealed the particles’ contents. That work led to a spinout called Cache DNA.One downside to that storage system is that it takes several days to embed DNA into the silica particles. Furthermore, removing the DNA from the particles requires hydrofluoric acid, which can be hazardous to workers handling the DNA.To come up with alternative storage materials, Banal began working with Johnson and members of his lab. Their idea was to use a type of polymer known as a degradable thermoset, which consists of polymers that form a solid when heated. The material also includes cleavable links that can be easily broken, allowing the polymer to be degraded in a controlled way.“With these deconstructable thermosets, depending on what cleavable bonds we put into them, we can choose how we want to degrade them,” Johnson says.For this project, the researchers decided to make their thermoset polymer from styrene and a cross-linker, which together form an amber-like thermoset called cross-linked polystyrene. This thermoset is also very hydrophobic, so it can prevent moisture from getting in and damaging the DNA. To make the thermoset degradable, the styrene monomers and cross-linkers are copolymerized with monomers called thionolactones. These links can be broken by treating them with a molecule called cysteamine.Because styrene is so hydrophobic, the researchers had to come up with a way to entice DNA — a hydrophilic, negatively charged molecule — into the styrene.To do that, they identified a combination of three monomers that they could turn into polymers that dissolve DNA by helping it interact with styrene. Each of the monomers has different features that cooperate to get the DNA out of water and into the styrene. There, the DNA forms spherical complexes, with charged DNA in the center and hydrophobic groups forming an outer layer that interacts with styrene. When heated, this solution becomes a solid glass-like block, embedded with DNA complexes.The researchers dubbed their method T-REX (Thermoset-REinforced Xeropreservation). The process of embedding DNA into the polymer network takes a few hours, but that could become shorter with further optimization, the researchers say.To release the DNA, the researchers first add cysteamine, which cleaves the bonds holding the polystyrene thermoset together, breaking it into smaller pieces. Then, a detergent called SDS can be added to remove the DNA from polystyrene without damaging it.Storing informationUsing these polymers, the researchers showed that they could encapsulate DNA of varying length, from tens of nucleotides up to an entire human genome (more than 50,000 base pairs). They were able to store DNA encoding the Emancipation Proclamation and the MIT logo, in addition to the theme music from “Jurassic Park.”After storing the DNA and then removing it, the researchers sequenced it and found that no errors had been introduced, which is a critical feature of any digital data storage system.The researchers also showed that the thermoset polymer can protect DNA from temperatures up to 75 degrees Celsius (167 degrees Fahrenheit). They are now working on ways to streamline the process of making the polymers and forming them into capsules for long-term storage.Cache DNA, a company started by Banal and Bathe, with Johnson as a member of the scientific advisory board, is now working on further developing DNA storage technology. The earliest application they envision is storing genomes for personalized medicine, and they also anticipate that these stored genomes could undergo further analysis as better technology is developed in the future.“The idea is, why don’t we preserve the master record of life forever?” Banal says. “Ten years or 20 years from now, when technology has advanced way more than we could ever imagine today, we could learn more and more things. We’re still in the very infancy of understanding the genome and how it relates to disease.”The research was funded by the National Science Foundation. More