Algorithms

Subterms

    More stories

    • 113 Shares149 Views

      in Data Management & Statistics

      3 Questions: Leo Anthony Celi on ChatGPT and medicine

      by

      Launched in November 2022, ChatGPT is a chatbot that can not only engage in human-like conversation, but also provide accurate answers to questions in a wide range of knowledge domains. The chatbot, created by the firm OpenAI, is based on a family of “large language models” — algorithms that can recognize, predict, and generate text based on patterns they identify in datasets containing hundreds of millions of words.

      In a study appearing in PLOS Digital Health this week, researchers report that ChatGPT performed at or near the passing threshold of the U.S. Medical Licensing Exam (USMLE) — a comprehensive, three-part exam that doctors must pass before practicing medicine in the United States. In an editorial accompanying the paper, Leo Anthony Celi, a principal research scientist at MIT’s Institute for Medical Engineering and Science, a practicing physician at Beth Israel Deaconess Medical Center, and an associate professor at Harvard Medical School, and his co-authors argue that ChatGPT’s success on this exam should be a wake-up call for the medical community.

      Q: What do you think the success of ChatGPT on the USMLE reveals about the nature of the medical education and evaluation of students? 

      A: The framing of medical knowledge as something that can be encapsulated into multiple choice questions creates a cognitive framing of false certainty. Medical knowledge is often taught as fixed model representations of health and disease. Treatment effects are presented as stable over time despite constantly changing practice patterns. Mechanistic models are passed on from teachers to students with little emphasis on how robustly those models were derived, the uncertainties that persist around them, and how they must be recalibrated to reflect advances worthy of incorporation into practice. 

      ChatGPT passed an examination that rewards memorizing the components of a system rather than analyzing how it works, how it fails, how it was created, how it is maintained. Its success demonstrates some of the shortcomings in how we train and evaluate medical students. Critical thinking requires appreciation that ground truths in medicine continually shift, and more importantly, an understanding how and why they shift.

      Q: What steps do you think the medical community should take to modify how students are taught and evaluated?  

      A: Learning is about leveraging the current body of knowledge, understanding its gaps, and seeking to fill those gaps. It requires being comfortable with and being able to probe the uncertainties. We fail as teachers by not teaching students how to understand the gaps in the current body of knowledge. We fail them when we preach certainty over curiosity, and hubris over humility.  

      Medical education also requires being aware of the biases in the way medical knowledge is created and validated. These biases are best addressed by optimizing the cognitive diversity within the community. More than ever, there is a need to inspire cross-disciplinary collaborative learning and problem-solving. Medical students need data science skills that will allow every clinician to contribute to, continually assess, and recalibrate medical knowledge.

      Q: Do you see any upside to ChatGPT’s success in this exam? Are there beneficial ways that ChatGPT and other forms of AI can contribute to the practice of medicine? 

      A: There is no question that large language models (LLMs) such as ChatGPT are very powerful tools in sifting through content beyond the capabilities of experts, or even groups of experts, and extracting knowledge. However, we will need to address the problem of data bias before we can leverage LLMs and other artificial intelligence technologies. The body of knowledge that LLMs train on, both medical and beyond, is dominated by content and research from well-funded institutions in high-income countries. It is not representative of most of the world.

      We have also learned that even mechanistic models of health and disease may be biased. These inputs are fed to encoders and transformers that are oblivious to these biases. Ground truths in medicine are continuously shifting, and currently, there is no way to determine when ground truths have drifted. LLMs do not evaluate the quality and the bias of the content they are being trained on. Neither do they provide the level of uncertainty around their output. But the perfect should not be the enemy of the good. There is tremendous opportunity to improve the way health care providers currently make clinical decisions, which we know are tainted with unconscious bias. I have no doubt AI will deliver its promise once we have optimized the data input. More

    • 100 Shares159 Views

      in Data Management & Statistics

      When should data scientists try a new technique?

      by

      If a scientist wanted to forecast ocean currents to understand how pollution travels after an oil spill, she could use a common approach that looks at currents traveling between 10 and 200 kilometers. Or, she could choose a newer model that also includes shorter currents. This might be more accurate, but it could also require learning new software or running new computational experiments. How to know if it will be worth the time, cost, and effort to use the new method?

      A new approach developed by MIT researchers could help data scientists answer this question, whether they are looking at statistics on ocean currents, violent crime, children’s reading ability, or any number of other types of datasets.

      The team created a new measure, known as the “c-value,” that helps users choose between techniques based on the chance that a new method is more accurate for a specific dataset. This measure answers the question “is it likely that the new method is more accurate for this data than the common approach?”

      Traditionally, statisticians compare methods by averaging a method’s accuracy across all possible datasets. But just because a new method is better for all datasets on average doesn’t mean it will actually provide a better estimate using one particular dataset. Averages are not application-specific.

      So, researchers from MIT and elsewhere created the c-value, which is a dataset-specific tool. A high c-value means it is unlikely a new method will be less accurate than the original method on a specific data problem.

      In their proof-of-concept paper, the researchers describe and evaluate the c-value using real-world data analysis problems: modeling ocean currents, estimating violent crime in neighborhoods, and approximating student reading ability at schools. They show how the c-value could help statisticians and data analysts achieve more accurate results by indicating when to use alternative estimation methods they otherwise might have ignored.

      “What we are trying to do with this particular work is come up with something that is data specific. The classical notion of risk is really natural for someone developing a new method. That person wants their method to work well for all of their users on average. But a user of a method wants something that will work on their individual problem. We’ve shown that the c-value is a very practical proof-of-concept in that direction,” says senior author Tamara Broderick, an associate professor in the Department of Electrical Engineering and Computer Science (EECS) and a member of the Laboratory for Information and Decision Systems and the Institute for Data, Systems, and Society.

      She’s joined on the paper by Brian Trippe PhD ’22, a former graduate student in Broderick’s group who is now a postdoc at Columbia University; and Sameer Deshpande ’13, a former postdoc in Broderick’s group who is now an assistant professor at the University of Wisconsin at Madison. An accepted version of the paper is posted online in the Journal of the American Statistical Association.

      Evaluating estimators

      The c-value is designed to help with data problems in which researchers seek to estimate an unknown parameter using a dataset, such as estimating average student reading ability from a dataset of assessment results and student survey responses. A researcher has two estimation methods and must decide which to use for this particular problem.

      The better estimation method is the one that results in less “loss,” which means the estimate will be closer to the ground truth. Consider again the forecasting of ocean currents: Perhaps being off by a few meters per hour isn’t so bad, but being off by many kilometers per hour makes the estimate useless. The ground truth is unknown, though; the scientist is trying to estimate it. Therefore, one can never actually compute the loss of an estimate for their specific data. That’s what makes comparing estimates challenging. The c-value helps a scientist navigate this challenge.

      The c-value equation uses a specific dataset to compute the estimate with each method, and then once more to compute the c-value between the methods. If the c-value is large, it is unlikely that the alternative method is going to be worse and yield less accurate estimates than the original method.

      “In our case, we are assuming that you conservatively want to stay with the default estimator, and you only want to go to the new estimator if you feel very confident about it. With a high c-value, it’s likely that the new estimate is more accurate. If you get a low c-value, you can’t say anything conclusive. You might have actually done better, but you just don’t know,” Broderick explains.

      Probing the theory

      The researchers put that theory to the test by evaluating three real-world data analysis problems.

      For one, they used the c-value to help determine which approach is best for modeling ocean currents, a problem Trippe has been tackling. Accurate models are important for predicting the dispersion of contaminants, like pollution from an oil spill. The team found that estimating ocean currents using multiple scales, one larger and one smaller, likely yields higher accuracy than using only larger scale measurements.

      “Oceans researchers are studying this, and the c-value can provide some statistical ‘oomph’ to support modeling the smaller scale,” Broderick says.

      In another example, the researchers sought to predict violent crime in census tracts in Philadelphia, an application Deshpande has been studying. Using the c-value, they found that one could get better estimates about violent crime rates by incorporating information about census-tract-level nonviolent crime into the analysis. They also used the c-value to show that additionally leveraging violent crime data from neighboring census tracts in the analysis isn’t likely to provide further accuracy improvements.

      “That doesn’t mean there isn’t an improvement, that just means that we don’t feel confident saying that you will get it,” she says.

      Now that they have proven the c-value in theory and shown how it could be used to tackle real-world data problems, the researchers want to expand the measure to more types of data and a wider set of model classes.

      The ultimate goal is to create a measure that is general enough for many more data analysis problems, and while there is still a lot of work to do to realize that objective, Broderick says this is an important and exciting first step in the right direction.

      This research was supported, in part, by an Advanced Research Projects Agency-Energy grant, a National Science Foundation CAREER Award, the Office of Naval Research, and the Wisconsin Alumni Research Foundation. More

    • 63 Shares129 Views

      in Data Management & Statistics

      Putting clear bounds on uncertainty

      by

      In science and technology, there has been a long and steady drive toward improving the accuracy of measurements of all kinds, along with parallel efforts to enhance the resolution of images. An accompanying goal is to reduce the uncertainty in the estimates that can be made, and the inferences drawn, from the data (visual or otherwise) that have been collected. Yet uncertainty can never be wholly eliminated. And since we have to live with it, at least to some extent, there is much to be gained by quantifying the uncertainty as precisely as possible.

      Expressed in other terms, we’d like to know just how uncertain our uncertainty is.

      That issue was taken up in a new study, led by Swami Sankaranarayanan, a postdoc at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), and his co-authors — Anastasios Angelopoulos and Stephen Bates of the University of California at Berkeley; Yaniv Romano of Technion, the Israel Institute of Technology; and Phillip Isola, an associate professor of electrical engineering and computer science at MIT. These researchers succeeded not only in obtaining accurate measures of uncertainty, they also found a way to display uncertainty in a manner the average person could grasp.

      Their paper, which was presented in December at the Neural Information Processing Systems Conference in New Orleans, relates to computer vision — a field of artificial intelligence that involves training computers to glean information from digital images. The focus of this research is on images that are partially smudged or corrupted (due to missing pixels), as well as on methods — computer algorithms, in particular — that are designed to uncover the part of the signal that is marred or otherwise concealed. An algorithm of this sort, Sankaranarayanan explains, “takes the blurred image as the input and gives you a clean image as the output” — a process that typically occurs in a couple of steps.

      First, there is an encoder, a kind of neural network specifically trained by the researchers for the task of de-blurring fuzzy images. The encoder takes a distorted image and, from that, creates an abstract (or “latent”) representation of a clean image in a form — consisting of a list of numbers — that is intelligible to a computer but would not make sense to most humans. The next step is a decoder, of which there are a couple of types, that are again usually neural networks. Sankaranarayanan and his colleagues worked with a kind of decoder called a “generative” model. In particular, they used an off-the-shelf version called StyleGAN, which takes the numbers from the encoded representation (of a cat, for instance) as its input and then constructs a complete, cleaned-up image (of that particular cat). So the entire process, including the encoding and decoding stages, yields a crisp picture from an originally muddied rendering.

      But how much faith can someone place in the accuracy of the resultant image? And, as addressed in the December 2022 paper, what is the best way to represent the uncertainty in that image? The standard approach is to create a “saliency map,” which ascribes a probability value — somewhere between 0 and 1 — to indicate the confidence the model has in the correctness of every pixel, taken one at a time. This strategy has a drawback, according to Sankaranarayanan, “because the prediction is performed independently for each pixel. But meaningful objects occur within groups of pixels, not within an individual pixel,” he adds, which is why he and his colleagues are proposing an entirely different way of assessing uncertainty.

      Their approach is centered around the “semantic attributes” of an image — groups of pixels that, when taken together, have meaning, making up a human face, for example, or a dog, or some other recognizable thing. The objective, Sankaranarayanan maintains, “is to estimate uncertainty in a way that relates to the groupings of pixels that humans can readily interpret.”

      Whereas the standard method might yield a single image, constituting the “best guess” as to what the true picture should be, the uncertainty in that representation is normally hard to discern. The new paper argues that for use in the real world, uncertainty should be presented in a way that holds meaning for people who are not experts in machine learning. Rather than producing a single image, the authors have devised a procedure for generating a range of images — each of which might be correct. Moreover, they can set precise bounds on the range, or interval, and provide a probabilistic guarantee that the true depiction lies somewhere within that range. A narrower range can be provided if the user is comfortable with, say, 90 percent certitude, and a narrower range still if more risk is acceptable.

      The authors believe their paper puts forth the first algorithm, designed for a generative model, which can establish uncertainty intervals that relate to meaningful (semantically-interpretable) features of an image and come with “a formal statistical guarantee.” While that is an important milestone, Sankaranarayanan considers it merely a step toward “the ultimate goal. So far, we have been able to do this for simple things, like restoring images of human faces or animals, but we want to extend this approach into more critical domains, such as medical imaging, where our ‘statistical guarantee’ could be especially important.”

      Suppose that the film, or radiograph, of a chest X-ray is blurred, he adds, “and you want to reconstruct the image. If you are given a range of images, you want to know that the true image is contained within that range, so you are not missing anything critical” — information that might reveal whether or not a patient has lung cancer or pneumonia. In fact, Sankaranarayanan and his colleagues have already begun working with a radiologist to see if their algorithm for predicting pneumonia could be useful in a clinical setting.

      Their work may also have relevance in the law enforcement field, he says. “The picture from a surveillance camera may be blurry, and you want to enhance that. Models for doing that already exist, but it is not easy to gauge the uncertainty. And you don’t want to make a mistake in a life-or-death situation.” The tools that he and his colleagues are developing could help identify a guilty person and help exonerate an innocent one as well.

      Much of what we do and many of the things happening in the world around us are shrouded in uncertainty, Sankaranarayanan notes. Therefore, gaining a firmer grasp of that uncertainty could help us in countless ways. For one thing, it can tell us more about exactly what it is we do not know.

      Angelopoulos was supported by the National Science Foundation. Bates was supported by the Foundations of Data Science Institute and the Simons Institute. Romano was supported by the Israel Science Foundation and by a Career Advancement Fellowship from Technion. Sankaranarayanan’s and Isola’s research for this project was sponsored by the U.S. Air Force Research Laboratory and the U.S. Air Force Artificial Intelligence Accelerator and was accomplished under Cooperative Agreement Number FA8750-19-2- 1000. MIT SuperCloud and the Lincoln Laboratory Supercomputing Center also provided computing resources that contributed to the results reported in this work. More

    • 113 Shares99 Views

      in Data Management & Statistics

      Research, education, and connection in the face of war

      by

      When Russian forces invaded Ukraine in February 2022, Tetiana Herasymova had several decisions to make: What should she do, where should she live, and should she take her MITx MicroMasters capstone exams? She had registered for the Statistics and Data Science Program’s final exams just days prior to moving out of her apartment and into a bomb shelter. Although it was difficult to focus on studying and preparations with air horns sounding overhead and uncertainty lingering around her, she was determined to try. “I wouldn’t let the aggressor in the war squash my dreams,” she says.

      A love of research and the desire to improve teaching 

      An early love of solving puzzles and problems for fun piqued Herasymova’s initial interest in mathematics. When she later pursued her PhD in mathematics at Kiev National Taras Shevchenko University, Herasymova’s love of math evolved into a love of research. Throughout Herasymova’s career, she’s worked to close the gap between scientific researchers and educators. Starting as a math tutor at MBA Strategy, a company that prepares Ukrainian leaders for qualifying standardized tests for MBA programs, she was later promoted as the head of their test preparation department. Afterward, she moved on to an equivalent position at ZNOUA, a new project that prepared high school students for Ukraine’s standardized test, and she eventually became ZNOUA’s CEO.

      In 2018, she founded Prosteer, a “self-learning community” of educators who share research, pedagogy, and experience to learn from one another. “It’s really interesting to have a community of teachers from different domains,” she says, speaking of educators and researchers whose specialties range across language, mathematics, physics, music, and more.

      Implementing new pedagogical research in the classroom is often up to educators who seek out studies on an individual basis, Herasymova has found. “Lots of scientists are not practitioners,” she says, and the reverse is also true. She only became more determined to build these connections once she was promoted to head of test preparation at MBA Strategy because she wanted to share more effective pedagogy with the tutors she was mentoring.

      First, Herasymova knew she needed a way to measure the teachers’ effectiveness. She was able to determine whether students who received the company’s tutoring services improved their scores. Moreover, Ukraine keeps an open-access database of national standardized test scores, so anyone could analyze the data in hopes of improving the level of education in the country. She says, “I could do some analytics because I am a mathematician, but I knew I could do much more with this data if I knew data science and machine learning knowledge.”

      That’s why Herasymova sought out the MITx MicroMasters Program in Statistics and Data Science offered by the MIT Institute for Data, Systems, and Society (IDSS). “I wanted to learn the fundamentals so I could join the Learning Analytics domain,” she says. She was looking for a comprehensive program that covered the foundations without being overly basic. “I had some knowledge from the ground, so I could see the deepness of that course,” she says. Because of her background as an instructional designer, she thought the MicroMasters curriculum was well-constructed, calling the variety of videos, practice problems, and homework assignments that encouraged learners to approach the course material in different ways, “a perfect experience.”

      Another benefit of the MicroMasters program was its online format. “I had my usual work, so it was impossible to study in a stationary way,” she says. She found the structure to be more flexible than other programs. “It’s really great that you can construct your course schedule your own way, especially with your own adult life,” she says.

      Determination and support in the midst of war

      When the war first forced Herasymova to flee her apartment, she had already registered to take the exams for her four courses. “It was quite hard to prepare for exams when you could hear explosions outside of the bomb shelter,” she says. She and other Ukranians were invited to postpone their exams until the following session, but the next available testing period wouldn’t be held until October. “It was a hard decision, but I had to allow myself to try,” she says. “For all people in Ukraine, when you don’t know if you’re going to live or die, you try to live in the now. You have to appreciate every moment and what life brings to you. You don’t say, ‘Someday’ — you do it today or tomorrow.”

      In addition to emotional support from her boyfriend, Herasymova had a group of friends who had also enrolled in the program, and they supported each other through study sessions and an ongoing chat. Herasymova’s personal support network helped her accomplish what she set out to do with her MicroMasters program, and in turn, she was able to support her professional network. While Prosteer halted its regular work during the early stages of the war, Herasymova was determined to support the community of educators and scientists that she had built. They continued meeting weekly to exchange ideas as usual. “It’s intrinsic motivation,” she says. They managed to restore all of their activities by October.

      Despite the factors stacked against her, Herasymova’s determination paid off — she passed all of her exams in May, the final step to earning her MicroMasters certificate in statistics and data science. “I just couldn’t believe it,” she says. “It was definitely a bifurcation point. The moment when you realize that you have something to rely on, and that life is just beginning to show all its diversity despite the fact that you live in war.” With her newly minted certificate in hand, Herasymova has continued her research on the effectiveness of educational models — analyzing the data herself — with a summer research program at New York University. 

      The student becomes the master

      After moving seven times between February and October, heading west from Kyiv until most recently settling near the border of Poland, Herasymova hopes she’s moved for the last time. Ukrainian Catholic University offered her a position teaching both mathematics and programming. Before enrolling in the MicroMasters Program in Statistics and Data Science, she had some prior knowledge of programming languages and mathematical algorithms, but she didn’t know Python. She took MITx’s Introduction to Computer Science and Programming Using Python to prepare. “It gave me a huge step forward,” she says. “I learned a lot. Now, not only can I work with Python machine learning models in programming language R, I also have knowledge of the big picture of the purpose and the point to do so.”

      In addition to the skills the MicroMasters Program trained her in, she gained firsthand experience in learning new subjects and exploring topics more deeply. She will be sharing that practice with the community of students and teachers she’s built, plus, she plans on guiding them through this course during the next year. As a continuation of her own educational growth, says she’s looking forward to her next MITx course this year, Data Analysis.

      Herasymova advises that the best way to keep progressing is investing a lot of time. “Adults don’t want to hear this, but you need one or two years,” she says. “Allow yourself to be stupid. If you’re an expert in one domain and want to switch to another, or if you want to understand something new, a lot of people don’t ask questions or don’t ask for help. But from this point, if I don’t know something, I know I should ask for help because that’s the start of learning. With a fixed mindset, you won’t grow.”

      July 2022 MicroMasters Program Joint Completion Celebration. Ukrainian student Tetiana Herasymova, who completed her program amid war in her home country, speaks at 43:55. More

    • 88 Shares169 Views

      in Data Management & Statistics

      Q&A: A fresh look at data science

      by

      As the leaders of a developing field, data scientists must often deal with a frustratingly slippery question: What is data science, precisely, and what is it good for?

      Alfred Spector is a visiting scholar in the MIT Department of Electrical Engineering and Computer Science (EECS), an influential developer of distributed computing systems and applications, and a successful tech executive with companies including IBM and Google. Along with three co-authors — Peter Norvig at Stanford University and Google, Chris Wiggins at Columbia University and The New York Times, and Jeannette M. Wing at Columbia — Spector recently published “Data Science in Context: Foundations, Challenges, Opportunities” (Cambridge University Press), which provides a broad, conversational overview of the wide-ranging field driving change in sectors ranging from health care to transportation to commerce to entertainment. 

      Here, Spector talks about data-driven life, what makes a good data scientist, and how his book came together during the height of the Covid-19 pandemic.

      Q: One of the most common buzzwords Americans hear is “data-driven,” but many might not know what that term is supposed to mean. Can you unpack it for us?

      A: Data-driven broadly refers to techniques or algorithms powered by data — they either provide insight or reach conclusions, say, a recommendation or a prediction. The algorithms power models which are increasingly woven into the fabric of science, commerce, and life, and they often provide excellent results. The list of their successes is really too long to even begin to list. However, one concern is that the proliferation of data makes it easy for us as students, scientists, or just members of the public to jump to erroneous conclusions. As just one example, our own confirmation biases make us prone to believing some data elements or insights “prove” something we already believe to be true. Additionally, we often tend to see causal relationships where the data only shows correlation. It might seem paradoxical, but data science makes critical reading and analysis of data all the more important.

      Q: What, to your mind, makes a good data scientist?

      A: [In talking to students and colleagues] I optimistically emphasize the power of data science and the importance of gaining the computational, statistical, and machine learning skills to apply it. But, I also remind students that we are obligated to solve problems well. In our book, Chris [Wiggins] paraphrases danah boyd, who says that a successful application of data science is not one that merely meets some technical goal, but one that actually improves lives. More specifically, I exhort practitioners to provide a real solution to problems, or else clearly identify what we are not solving so that people see the limitations of our work. We should be extremely clear so that we do not generate harmful results or lead others to erroneous conclusions. I also remind people that all of us, including scientists and engineers, are human and subject to the same human foibles as everyone else, such as various biases. 

      Q: You discuss Covid-19 in your book. While some short-range models for mortality were very accurate during the heart of the pandemic, you note the failure of long-range models to predict any of 2020’s four major geotemporal Covid waves in the United States. Do you feel Covid was a uniquely hard situation to model? 

      A: Covid was particularly difficult to predict over the long term because of many factors — the virus was changing, human behavior was changing, political entities changed their minds. Also, we didn’t have fine-grained mobility data (perhaps, for good reasons), and we lacked sufficient scientific understanding of the virus, particularly in the first year.

      I think there are many other domains which are similarly difficult. Our book teases out many reasons why data-driven models may not be applicable. Perhaps it’s too difficult to get or hold the necessary data. Perhaps the past doesn’t predict the future. If data models are being used in life-and-death situations, we may not be able to make them sufficiently dependable; this is particularly true as we’ve seen all the motivations that bad actors have to find vulnerabilities. So, as we continue to apply data science, we need to think through all the requirements we have, and the capability of the field to meet them. They often align, but not always. And, as data science seeks to solve problems into ever more important areas such as human health, education, transportation safety, etc., there will be many challenges.

      Q: Let’s talk about the power of good visualization. You mention the popular, early 2000’s Baby Name Voyager website as one that changed your view on the importance of data visualization. Tell us how that happened. 

      A: That website, recently reborn as the Name Grapher, had two characteristics that I thought were brilliant. First, it had a really natural interface, where you type the initial characters of a name and it shows a frequency graph of all the names beginning with those letters, and their popularity over time. Second, it’s so much better than a spreadsheet with 140 columns representing years and rows representing names, despite the fact it contains no extra information. It also provided instantaneous feedback with its display graph dynamically changing as you type. To me, this showed the power of a very simple transformation that is done correctly.

      Q: When you and your co-authors began planning “Data Science In Context,” what did you hope to offer?

      A: We portray present data science as a field that’s already had enormous benefits, that provides even more future opportunities, but one that requires equally enormous care in its use. Referencing the word “context” in the title, we explain that the proper use of data science must consider the specifics of the application, the laws and norms of the society in which the application is used, and even the time period of its deployment. And, importantly for an MIT audience, the practice of data science must go beyond just the data and the model to the careful consideration of an application’s objectives, its security, privacy, abuse, and resilience risks, and even the understandability it conveys to humans. Within this expansive notion of context, we finally explain that data scientists must also carefully consider ethical trade-offs and societal implications.

      Q: How did you keep focus throughout the process?

      A: Much like in open-source projects, I played both the coordinating author role and also the role of overall librarian of all the material, but we all made significant contributions. Chris Wiggins is very knowledgeable on the Belmont principles and applied ethics; he was the major contributor of those sections. Peter Norvig, as the coauthor of a bestselling AI textbook, was particularly involved in the sections on building models and causality. Jeannette Wing worked with me very closely on our seven-element Analysis Rubric and recognized that a checklist for data science practitioners would end up being one of our book’s most important contributions. 

      From a nuts-and-bolts perspective, we wrote the book during Covid, using one large shared Google doc with weekly video conferences. Amazingly enough, Chris, Jeannette, and I didn’t meet in person at all, and Peter and I met only once — sitting outdoors on a wooden bench on the Stanford campus.

      Q: That is an unusual way to write a book! Do you recommend it?

      A: It would be nice to have had more social interaction, but a shared document, at least with a coordinating author, worked pretty well for something up to this size. The benefit is that we always had a single, coherent textual base, not dissimilar to how a programming team works together.

      This is a condensed, edited version of a longer interview that originally appeared on the MIT EECS website. More

    • 113 Shares109 Views

      in Data Management & Statistics

      Unpacking the “black box” to build better AI models

      by

      When deep learning models are deployed in the real world, perhaps to detect financial fraud from credit card activity or identify cancer in medical images, they are often able to outperform humans.

      But what exactly are these deep learning models learning? Does a model trained to spot skin cancer in clinical images, for example, actually learn the colors and textures of cancerous tissue, or is it flagging some other features or patterns?

      These powerful machine-learning models are typically based on artificial neural networks that can have millions of nodes that process data to make predictions. Due to their complexity, researchers often call these models “black boxes” because even the scientists who build them don’t understand everything that is going on under the hood.

      Stefanie Jegelka isn’t satisfied with that “black box” explanation. A newly tenured associate professor in the MIT Department of Electrical Engineering and Computer Science, Jegelka is digging deep into deep learning to understand what these models can learn and how they behave, and how to build certain prior information into these models.

      “At the end of the day, what a deep-learning model will learn depends on so many factors. But building an understanding that is relevant in practice will help us design better models, and also help us understand what is going on inside them so we know when we can deploy a model and when we can’t. That is critically important,” says Jegelka, who is also a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL) and the Institute for Data, Systems, and Society (IDSS).

      Jegelka is particularly interested in optimizing machine-learning models when input data are in the form of graphs. Graph data pose specific challenges: For instance, information in the data consists of both information about individual nodes and edges, as well as the structure — what is connected to what. In addition, graphs have mathematical symmetries that need to be respected by the machine-learning model so that, for instance, the same graph always leads to the same prediction. Building such symmetries into a machine-learning model is usually not easy.

      Take molecules, for instance. Molecules can be represented as graphs, with vertices that correspond to atoms and edges that correspond to chemical bonds between them. Drug companies may want to use deep learning to rapidly predict the properties of many molecules, narrowing down the number they must physically test in the lab.

      Jegelka studies methods to build mathematical machine-learning models that can effectively take graph data as an input and output something else, in this case a prediction of a molecule’s chemical properties. This is particularly challenging since a molecule’s properties are determined not only by the atoms within it, but also by the connections between them.  

      Other examples of machine learning on graphs include traffic routing, chip design, and recommender systems.

      Designing these models is made even more difficult by the fact that data used to train them are often different from data the models see in practice. Perhaps the model was trained using small molecular graphs or traffic networks, but the graphs it sees once deployed are larger or more complex.

      In this case, what can researchers expect this model to learn, and will it still work in practice if the real-world data are different?

      “Your model is not going to be able to learn everything because of some hardness problems in computer science, but what you can learn and what you can’t learn depends on how you set the model up,” Jegelka says.

      She approaches this question by combining her passion for algorithms and discrete mathematics with her excitement for machine learning.

      From butterflies to bioinformatics

      Jegelka grew up in a small town in Germany and became interested in science when she was a high school student; a supportive teacher encouraged her to participate in an international science competition. She and her teammates from the U.S. and Singapore won an award for a website they created about butterflies, in three languages.

      “For our project, we took images of wings with a scanning electron microscope at a local university of applied sciences. I also got the opportunity to use a high-speed camera at Mercedes Benz — this camera usually filmed combustion engines — which I used to capture a slow-motion video of the movement of a butterfly’s wings. That was the first time I really got in touch with science and exploration,” she recalls.

      Intrigued by both biology and mathematics, Jegelka decided to study bioinformatics at the University of Tübingen and the University of Texas at Austin. She had a few opportunities to conduct research as an undergraduate, including an internship in computational neuroscience at Georgetown University, but wasn’t sure what career to follow.

      When she returned for her final year of college, Jegelka moved in with two roommates who were working as research assistants at the Max Planck Institute in Tübingen.

      “They were working on machine learning, and that sounded really cool to me. I had to write my bachelor’s thesis, so I asked at the institute if they had a project for me. I started working on machine learning at the Max Planck Institute and I loved it. I learned so much there, and it was a great place for research,” she says.

      She stayed on at the Max Planck Institute to complete a master’s thesis, and then embarked on a PhD in machine learning at the Max Planck Institute and the Swiss Federal Institute of Technology.

      During her PhD, she explored how concepts from discrete mathematics can help improve machine-learning techniques.

      Teaching models to learn

      The more Jegelka learned about machine learning, the more intrigued she became by the challenges of understanding how models behave, and how to steer this behavior.

      “You can do so much with machine learning, but only if you have the right model and data. It is not just a black-box thing where you throw it at the data and it works. You actually have to think about it, its properties, and what you want the model to learn and do,” she says.

      After completing a postdoc at the University of California at Berkeley, Jegelka was hooked on research and decided to pursue a career in academia. She joined the faculty at MIT in 2015 as an assistant professor.

      “What I really loved about MIT, from the very beginning, was that the people really care deeply about research and creativity. That is what I appreciate the most about MIT. The people here really value originality and depth in research,” she says.

      That focus on creativity has enabled Jegelka to explore a broad range of topics.

      In collaboration with other faculty at MIT, she studies machine-learning applications in biology, imaging, computer vision, and materials science.

      But what really drives Jegelka is probing the fundamentals of machine learning, and most recently, the issue of robustness. Often, a model performs well on training data, but its performance deteriorates when it is deployed on slightly different data. Building prior knowledge into a model can make it more reliable, but understanding what information the model needs to be successful and how to build it in is not so simple, she says.

      She is also exploring methods to improve the performance of machine-learning models for image classification.

      Image classification models are everywhere, from the facial recognition systems on mobile phones to tools that identify fake accounts on social media. These models need massive amounts of data for training, but since it is expensive for humans to hand-label millions of images, researchers often use unlabeled datasets to pretrain models instead.

      These models then reuse the representations they have learned when they are fine-tuned later for a specific task.

      Ideally, researchers want the model to learn as much as it can during pretraining, so it can apply that knowledge to its downstream task. But in practice, these models often learn only a few simple correlations — like that one image has sunshine and one has shade — and use these “shortcuts” to classify images.

      “We showed that this is a problem in ‘contrastive learning,’ which is a standard technique for pre-training, both theoretically and empirically. But we also show that you can influence the kinds of information the model will learn to represent by modifying the types of data you show the model. This is one step toward understanding what models are actually going to do in practice,” she says.

      Researchers still don’t understand everything that goes on inside a deep-learning model, or details about how they can influence what a model learns and how it behaves, but Jegelka looks forward to continue exploring these topics.

      “Often in machine learning, we see something happen in practice and we try to understand it theoretically. This is a huge challenge. You want to build an understanding that matches what you see in practice, so that you can do better. We are still just at the beginning of understanding this,” she says.

      Outside the lab, Jegelka is a fan of music, art, traveling, and cycling. But these days, she enjoys spending most of her free time with her preschool-aged daughter. More

    • 100 Shares149 Views

      in Data Management & Statistics

      Subtle biases in AI can influence emergency decisions

      by

      It’s no secret that people harbor biases — some unconscious, perhaps, and others painfully overt. The average person might suppose that computers — machines typically made of plastic, steel, glass, silicon, and various metals — are free of prejudice. While that assumption may hold for computer hardware, the same is not always true for computer software, which is programmed by fallible humans and can be fed data that is, itself, compromised in certain respects.

      Artificial intelligence (AI) systems — those based on machine learning, in particular — are seeing increased use in medicine for diagnosing specific diseases, for example, or evaluating X-rays. These systems are also being relied on to support decision-making in other areas of health care. Recent research has shown, however, that machine learning models can encode biases against minority subgroups, and the recommendations they make may consequently reflect those same biases.

      A new study by researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and the MIT Jameel Clinic, which was published last month in Communications Medicine, assesses the impact that discriminatory AI models can have, especially for systems that are intended to provide advice in urgent situations. “We found that the manner in which the advice is framed can have significant repercussions,” explains the paper’s lead author, Hammaad Adam, a PhD student at MIT’s Institute for Data Systems and Society. “Fortunately, the harm caused by biased models can be limited (though not necessarily eliminated) when the advice is presented in a different way.” The other co-authors of the paper are Aparna Balagopalan and Emily Alsentzer, both PhD students, and the professors Fotini Christia and Marzyeh Ghassemi.

      AI models used in medicine can suffer from inaccuracies and inconsistencies, in part because the data used to train the models are often not representative of real-world settings. Different kinds of X-ray machines, for instance, can record things differently and hence yield different results. Models trained predominately on white people, moreover, may not be as accurate when applied to other groups. The Communications Medicine paper is not focused on issues of that sort but instead addresses problems that stem from biases and on ways to mitigate the adverse consequences.

      A group of 954 people (438 clinicians and 516 nonexperts) took part in an experiment to see how AI biases can affect decision-making. The participants were presented with call summaries from a fictitious crisis hotline, each involving a male individual undergoing a mental health emergency. The summaries contained information as to whether the individual was Caucasian or African American and would also mention his religion if he happened to be Muslim. A typical call summary might describe a circumstance in which an African American man was found at home in a delirious state, indicating that “he has not consumed any drugs or alcohol, as he is a practicing Muslim.” Study participants were instructed to call the police if they thought the patient was likely to turn violent; otherwise, they were encouraged to seek medical help.

      The participants were randomly divided into a control or “baseline” group plus four other groups designed to test responses under slightly different conditions. “We want to understand how biased models can influence decisions, but we first need to understand how human biases can affect the decision-making process,” Adam notes. What they found in their analysis of the baseline group was rather surprising: “In the setting we considered, human participants did not exhibit any biases. That doesn’t mean that humans are not biased, but the way we conveyed information about a person’s race and religion, evidently, was not strong enough to elicit their biases.”

      The other four groups in the experiment were given advice that either came from a biased or unbiased model, and that advice was presented in either a “prescriptive” or a “descriptive” form. A biased model would be more likely to recommend police help in a situation involving an African American or Muslim person than would an unbiased model. Participants in the study, however, did not know which kind of model their advice came from, or even that models delivering the advice could be biased at all. Prescriptive advice spells out what a participant should do in unambiguous terms, telling them they should call the police in one instance or seek medical help in another. Descriptive advice is less direct: A flag is displayed to show that the AI system perceives a risk of violence associated with a particular call; no flag is shown if the threat of violence is deemed small.  

      A key takeaway of the experiment is that participants “were highly influenced by prescriptive recommendations from a biased AI system,” the authors wrote. But they also found that “using descriptive rather than prescriptive recommendations allowed participants to retain their original, unbiased decision-making.” In other words, the bias incorporated within an AI model can be diminished by appropriately framing the advice that’s rendered. Why the different outcomes, depending on how advice is posed? When someone is told to do something, like call the police, that leaves little room for doubt, Adam explains. However, when the situation is merely described — classified with or without the presence of a flag — “that leaves room for a participant’s own interpretation; it allows them to be more flexible and consider the situation for themselves.”

      Second, the researchers found that the language models that are typically used to offer advice are easy to bias. Language models represent a class of machine learning systems that are trained on text, such as the entire contents of Wikipedia and other web material. When these models are “fine-tuned” by relying on a much smaller subset of data for training purposes — just 2,000 sentences, as opposed to 8 million web pages — the resultant models can be readily biased.  

      Third, the MIT team discovered that decision-makers who are themselves unbiased can still be misled by the recommendations provided by biased models. Medical training (or the lack thereof) did not change responses in a discernible way. “Clinicians were influenced by biased models as much as non-experts were,” the authors stated.

      “These findings could be applicable to other settings,” Adam says, and are not necessarily restricted to health care situations. When it comes to deciding which people should receive a job interview, a biased model could be more likely to turn down Black applicants. The results could be different, however, if instead of explicitly (and prescriptively) telling an employer to “reject this applicant,” a descriptive flag is attached to the file to indicate the applicant’s “possible lack of experience.”

      The implications of this work are broader than just figuring out how to deal with individuals in the midst of mental health crises, Adam maintains.  “Our ultimate goal is to make sure that machine learning models are used in a fair, safe, and robust way.” More

    • 63 Shares149 Views

      in Data Management & Statistics

      Meet the 2022-23 Accenture Fellows

      by

      Launched in October 2020, the MIT and Accenture Convergence Initiative for Industry and Technology underscores the ways in which industry and technology can collaborate to spur innovation. The five-year initiative aims to achieve its mission through research, education, and fellowships. To that end, Accenture has once again awarded five annual fellowships to MIT graduate students working on research in industry and technology convergence who are underrepresented, including by race, ethnicity, and gender.This year’s Accenture Fellows work across research areas including telemonitoring, human-computer interactions, operations research,  AI-mediated socialization, and chemical transformations. Their research covers a wide array of projects, including designing low-power processing hardware for telehealth applications; applying machine learning to streamline and improve business operations; improving mental health care through artificial intelligence; and using machine learning to understand the environmental and health consequences of complex chemical reactions.As part of the application process, student nominations were invited from each unit within the School of Engineering, as well as from the Institute’s four other schools and the MIT Schwarzman College of Computing. Five exceptional students were selected as fellows for the initiative’s third year.Drew Buzzell is a doctoral candidate in electrical engineering and computer science whose research concerns telemonitoring, a fast-growing sphere of telehealth in which information is collected through internet-of-things (IoT) connected devices and transmitted to the cloud. Currently, the high volume of information involved in telemonitoring — and the time and energy costs of processing it — make data analysis difficult. Buzzell’s work is focused on edge computing, a new computing architecture that seeks to address these challenges by managing data closer to the source, in a distributed network of IoT devices. Buzzell earned his BS in physics and engineering science and his MS in engineering science from the Pennsylvania State University.

      Mengying (Cathy) Fang is a master’s student in the MIT School of Architecture and Planning. Her research focuses on augmented reality and virtual reality platforms. Fang is developing novel sensors and machine components that combine computation, materials science, and engineering. Moving forward, she will explore topics including soft robotics techniques that could be integrated with clothes and wearable devices and haptic feedback in order to develop interactions with digital objects. Fang earned a BS in mechanical engineering and human-computer interaction from Carnegie Mellon University.

      Xiaoyue Gong is a doctoral candidate in operations research at the MIT Sloan School of Management. Her research aims to harness the power of machine learning and data science to reduce inefficiencies in the operation of businesses, organizations, and society. With the support of an Accenture Fellowship, Gong seeks to find solutions to operational problems by designing reinforcement learning methods and other machine learning techniques to embedded operational problems. Gong earned a BS in honors mathematics and interactive media arts from New York University.

      Ruby Liu is a doctoral candidate in medical engineering and medical physics. Their research addresses the growing pandemic of loneliness among older adults, which leads to poor health outcomes and presents particularly high risks for historically marginalized people, including members of the LGBTQ+ community and people of color. Liu is designing a network of interconnected AI agents that foster connections between user and agent, offering mental health care while strengthening and facilitating human-human connections. Liu received a BS in biomedical engineering from Johns Hopkins University.

      Joules Provenzano is a doctoral candidate in chemical engineering. Their work integrates machine learning and liquid chromatography-high resolution mass spectrometry (LC-HRMS) to improve our understanding of complex chemical reactions in the environment. As an Accenture Fellow, Provenzano will build upon recent advances in machine learning and LC-HRMS, including novel algorithms for processing real, experimental HR-MS data and new approaches in extracting structure-transformation rules and kinetics. Their research could speed the pace of discovery in the chemical sciences and benefits industries including oil and gas, pharmaceuticals, and agriculture. Provenzano earned a BS in chemical engineering and international and global studies from the Rochester Institute of Technology. More