Mitigating hazards with vulnerability in mind

From tropical storms to landslides, the form and frequency of natural hazards vary widely. But the feelings of vulnerability they can provoke are universal.

Growing up in hazard-prone cities, Ipek Bensu Manav, a civil and environmental engineering PhD candidate with the MIT Concrete Sustainability Hub (CSHub), noticed that this vulnerability was always at the periphery. Today, she’s studying vulnerability, in both its engineering and social dimensions, with the aim of promoting more hazard-resilient communities.

Her research at CSHub has taken her across the country to attend impactful conferences and allowed her to engage with prominent experts and decision-makers in the realm of resilience. But more fundamentally, it has also taken her beyond the conventional bounds of engineering, reshaping her understanding of the practice.

From her time in Miami, Florida, and Istanbul, Turkey, Manav is no stranger to natural hazards. Istanbul, which suffered a devastating earthquake in 1999, is predicted to experience an equally violent tremor in the near future, while Miami ranks among the top cities in the U.S. in terms of natural disaster risk due to its vulnerability to hurricanes.

“Growing up in Miami, I’d always hear about hurricane season on the news,” recounts Manav, “While in Istanbul there was a constant fear about the next big earthquake. Losing people and [witnessing] those kinds of events instilled in me a desire to tame nature.”

It was this desire to “push the bounds of what is possible” — and to protect lives in the process — that motivated Manav to study civil engineering at Boğaziçi University. Her studies there affirmed her belief in the formidable power of engineering to “outsmart nature.”

This, in part, led her to continue her studies at MIT CSHub — a team of interdisciplinary researchers who study how to achieve resilient and sustainable infrastructure. Her role at CSHub has given her the opportunity to study resilience in depth. It has also challenged her understanding of natural disasters — and whether they are “natural” at all.

“Over the past few decades, some policy choices have increased the risk of experiencing disasters,” explains Manav. “An increasingly popular sentiment among resilience researchers is that natural disasters are not ‘natural,’ but are actually man-made. At CSHub we believe there is an opportunity to do better with the growing knowledge and engineering and policy research.”

As a part of the CSHub portfolio, Manav’s research looks not just at resilient engineering, but the engineering of resilient communities.

Her work draws on a metric developed at CSHub known as city texture, which is a measurement of the rectilinearity of a city’s layout. City texture, Manav and her colleagues have found, is a versatile and informative measurement. By capturing a city’s order or disorder, it can predict variations in wind flow — variations currently too computationally intensive for most cities to easily render.  

Manav has derived this metric for her native South Florida. A city texture analysis she conducted there found that numerous census tracts could experience wind speeds 50 percent greater than currently predicted. Mitigating these wind variations could lead to some $697 million in savings annually.

Such enormous hazard losses and the growing threat of climate change have presented her with a new understanding of engineering.

“With resilience and climate change at the forefront of engineering, the focus has shifted,” she explains, “from defying limits and building impressive structures to making structures that adapt to the changing environment around us.”

Witnessing this shift has reoriented her relationship with engineering. Rather than viewing it as a distinct science, she has begun to place it in its broader social and political context — and to recognize how those social and political dynamics often determine engineering outcomes.

“When I started grad school, I often felt ‘Oh this is an engineering problem. I can engineer a solution’,” recounts Manav. “But as I’ve read more about resilience, I’ve realized that it’s just as much a concern of politics and policy as it is of engineering.”

She attributes her awareness of policy to MIT CSHub’s collaboration with the Portland Cement Association and the Ready Mixed Concrete Research & Education Foundation. The commitment of the concrete and cement industries to resilient construction has exposed her to the myriad policies that dictate the resilience of communities.

“Spending time with our partners made me realize how much of a policy issue [resilience] is,” she explains. “And working with them has provided me with a seat at the table with the people engaged in resilience.”

Opportunities for engagement have been plentiful. She has attended numerous conferences and met with leaders in the realm of sustainability and resilience, including the International Code Council (ICC), Smart Home America, and Strengthen Alabama Homes.

Some opportunities have proven particularly fortuitous. When attending a presentation hosted by the ICC and the National Association for the Advancement of Colored People (NAACP) that highlighted people of color working on building codes, Manav felt inspired to reach out to the presenters. Soon after, she found herself collaborating with them on a policy report on resilience in communities of color.

“For me, it was a shifting point, going from prophesizing about what we could be doing, to observing what is being done. It was a very humbling experience,” she says. “Having worked in this lab made me feel more comfortable stepping outside of my comfort zone and reaching out.”

Manav credits this growing confidence to her mentorship at CSHub. More than just providing support, CSHub Co-director Randy Kirchain has routinely challenged her and inspired further growth.

“There have been countless times that I’ve reached out to him because I was feeling unsure of myself or my ideas,” says Manav. “And he’s offered clarity and assurance.”

Before her first conference, she recalls Kirchain staying in the office well into the evening to help her practice and hone her presentation. He’s also advocated for her on research projects to ensure that her insight is included and that she receives the credit she deserves. But most of all, he’s been a great person to work with.

“Randy is a lighthearted, funny, and honest person to be around,” recounts Manav. “He builds in me the confidence to dive straight into whatever task I’m tackling.”

That current task is related to equity. Inspired by her conversations with members of the NAACP, Manav has introduced a new dimension to her research — social vulnerability.

In contrast to place vulnerability, which captures the geographical susceptibility to hazards, social vulnerability captures the extent to which residents have the resources to respond to and recover from hazard events. Household income could act as a proxy for these resources, and the spread of household income across geographies and demographics can help derive metrics of place and social vulnerability. And these metrics matter.

“Selecting different metrics favors different people when distributing hazard mitigation and recovery funds,” explains Manav. “If we’re looking at just the dollar value of losses, then wealthy households with more valuable properties disproportionally benefit. But, conversely, if we look at losses as a percentage of income, we’re going to prioritize low-income households that might not necessarily have the resources to recover.”

Manav has incorporated metrics of social vulnerability into her city texture loss estimations. The resulting approach could predict unmitigated damage, estimate subsequent hazard losses, and measure the disparate impact of those losses on low-income and socially vulnerable communities.

Her hope is that this streamlined approach could change how funds are disbursed and give communities the tools to solve the entwined challenges of climate change and equity.

The city texture work Manav has adopted is quite different from the gravity-defying engineering that drew her to the field. But she’s found that it is often more pragmatic and impactful.

Rather than mastering the elements, she’s learning how to adapt to them and help others do the same. Solutions to climate change, she’s discovered, demand the collaboration of numerous parties — as well as a willingness to confront one’s own vulnerabilities and make the decision to reach out. 

de MIT News https://ift.tt/3taXRVs

J-WAFS announces 2021 Solutions Grants for commercializing water and food technologies

The Abdul Latif Jameel Water and Food Systems Lab (J-WAFS) recently announced the 2021 J-WAFS Solutions grant recipients. The J-WAFS Solutions program aims to propel MIT water- and food-related research toward commercialization. Grant recipients receive one year of financial support, as well as mentorship, networking, and guidance from industry experts, to begin their journey into the commercial world — whether that be in the form of bringing innovative products to market or launching cutting-edge startup companies. 

This year, three projects will receive funding across water, food, and agriculture spaces. The winning projects will advance nascent technologies for off-grid refrigeration, portable water filtration, and dairy waste recycling. Each provides an efficient, accessible solution to the respective challenge being addressed.

Since the start of the J-WAFS Solutions program in 2015, grants have provided instrumental support in creating a number of key MIT startups that focus on major water and food challenges. A 2015-16 grant helped the team behind Via Separations develop their business plan to massively decarbonize industrial separations processes. Other successful J-WAFS Solutions alumni include researchers who created a low-cost water filter made from tree branches and the team that launched the startup Xibus Systems, which is developing a handheld food safety sensor.

“New technological advances are being made at MIT every day, and J-WAFS Solutions grants provide critical resources and support for these technologies to make it to market so that they can transform our local and global water and food systems,” says J-WAFS Executive Director Renee Robins. “This year’s grant recipients offer innovative tools that will provide more accessible food storage for smallholder farmers in places like Africa, safer drinking water, and a new approach to recycling food waste,” Robins notes. She adds, “J-WAFS is excited to work with these teams, and we look forward to seeing their impact on the water and food sectors.”

The J-WAFS Solutions program is implemented in collaboration with Community Jameel, the global philanthropic organization founded by MIT Mohammed Jameel, and is supported by the MIT Venture Mentoring Service and the iCorps New England Regional Innovation Node at MIT.

Mobile evaporative cooling rooms for vegetable preservation

Food waste is a persistent problem across food systems supply chains, as 30-50 percent of food produced is lost before it reaches the table. The problem is compounded in areas without access to the refrigeration necessary to store food after it is harvested. Hot and dry climates in particular struggle to preserve food before it reaches consumers. A team led by Daniel Frey, faculty director for research at MIT D-Lab and professor of mechanical engineering, has pioneered a new approach to enable farmers to better preserve their produce and improve access to nutritious food in the community. The team includes Leon Glicksman, professor of building technology and mechanical engineering, and Eric Verploegen, a research engineer in MIT D-Lab.

Instead of relying on traditional refrigeration with high energy and cost requirements, the team is utilizing forced-air evaporative cooling chambers. Their design, based on retrofitting shipping containers, will provide a lower-cost, better-performing solution enabling farmers to chill their produce without access to power. The research team was previously funded by J-WAFS through two different grants in 2019 to develop the off-grid technology in collaboration with researchers at the University of Nairobi and the Collectives for Integrated Livelihood Initiatives (CInI), Jamshedpur. Now, the cooling rooms are ready for pilot testing, which the MIT team will conduct with rural farmers in Kenya and India. The MIT team will deploy and test the storage chambers through collaborations with two Kenyan social enterprises and a nongovernmental organization in Gujarat, India. 

Off-grid portable ion concentration polarization desalination unit

Shrinking aquifers, polluted rivers, and increased drought are making fresh drinking water increasingly scarce, driving the need for improved desalination technologies. The water purifiers market, which was $45 billion in 2019, is expected to grow to $90.1 billion in 2025. However, current products on the market are limited in scope, in that they are designed to treat water that is already relatively low in salinity, and do not account for lead contamination or other technical challenges. A better solution is required to ensure access to clean and safe drinking water in the face of water shortages. 

A team led by Jongyoon Han, professor of biological engineering and electrical engineering at MIT, has developed a portable desalination unit that utilizes an ion concentration polarization process. The compact and lightweight unit has the ability to remove dissolved and suspended solids from brackish water at a rate of one liter per hour, both in installed and remote field settings. The unit was featured in an award-winning video in the 2021 J-WAFS World Water Day Video Competition: MIT Research for a Water Secure Future. The team plans to develop the next-generation prototype of the desalination unit alongside a mass-production strategy and business model.

Converting dairy industry waste into food and feed ingredients

One of the trendiest foods in the last decade, Greek yogurt, has a hidden dark side: acid whey. This low-pH, liquid by-product of yogurt production has been a growing problem for producers, as untreated disposal of the whey can pose environmental risks due to its high organic content and acidic odor.

With an estimated 3 million tons of acid whey generated in the United States each year, MIT researchers saw an opportunity to turn waste into a valuable resource for our food systems. Led by the Willard Henry Dow Professor in Chemical Engineering, Gregory Stephanopoulos, and Anthony J. Sinskey, professor of microbiology, the researchers are utilizing metabolic engineering to turn acid whey into carotenoids, the yellow and orange organic pigments found naturally in carrots, autumn leaves, and salmon. The team is hoping that these carotenoids can be utilized as food supplements or feed additives to make the most of what otherwise would have been wasted.

de MIT News https://ift.tt/3DHfwJm

Helping underrepresented doctoral students of color thrive in the broader MIT community

The MIT University Center for Exemplary Mentoring (UCEM) was founded in 2015 with an Alfred P. Sloan Foundation grant that centers on the recruitment, retention, and academic success of underrepresented doctoral students of color in five areas within the School of Engineering: the departments of Biological Engineering, Chemical Engineering, Electrical Engineering and Computer Science (EECS), and Mechanical Engineering, and the Institute for Medical Engineering and Science/Harvard-MIT Program in Health Sciences and Technology.

Promising PhD candidates are recruited for the UCEM program, and once enrolled receive financial support, mentorship, and professional development training as well as access to a broad and diverse professional network. 

UCEM engages candidates through the entire process of “thinking about a PhD, getting in and doing a PhD, and then exiting the PhD,” says Leslie Kolodziejski, UCEM principal investigator and professor of electrical engineering in the Department of Electrical Engineering and Computer Science (EECS). Over the life of UCEM thus far, 83 scholars have been supported, with 58 scholars currently in the program.

A valued space at MIT

UCEM Scholar Jean Carlos Serrano Flores SM ’18 PhD ’21, a graduate of the Department of Mechanical Engineering, remembers first encountering UCEM at an MIT event before he had formally accepted his offer of admission.

“I had two students from [UCEM] directly talk to me prior to my decision,” Serrano Flores says. “UCEM made me feel much more secure that I was going someplace where I’d be welcome … and supported.”

Danielle Olson ’14, SM ’19, PhD ’21, also a UCEM scholar and a graduate of EECS, names the UCEM funding as being an important part of her choice to attend MIT. She also said that she’s appreciated the flexibility of the funding, which can be used to attend conferences and for research expenses that aren’t covered under lab funding.

But both Olson and Serrano Flores emphasize the value of the community and mentorship elements of the UCEM program. UCEM scholars get to know each other through seminars, one-on-one mentoring, conferences, and other UCEM events, such as regular scholar lunches and social activities.

Serrano Flores describes seminars ranging from developing different aspects of a scholar’s academic career to navigating interpersonal relationships with advisors and peers to managing racial biases that graduate students of color experience in academia.

“We're all from underrepresented backgrounds,” says Serrano Flores. “Being in a room with everybody that knows the experiences that you would go [through] along the way, and how tricky it is to really manage them in today's world, was really good.”

Olson feels similarly.

“I felt certainly firsthand in my first several years of graduate school that it's really easy to feel like you could fall between the cracks if you don't have support in place,” she says.

Olson says she was able to access one-on-one support through UCEM, in addition to finding space to talk about common challenges such as “imposter syndrome.”

“I wouldn't necessarily go to a network event in my community of research to talk about those things,” she says. “But [at UCEM] you have these spaces where you can not only have honest conversations … but also get strategies for addressing those things.”

Bianca Lepe is a fourth-year PhD candidate in the Department of Biological Engineering, a UCEM scholar, and the Graduate Student Council’s diversity, equity, and inclusion chair. She says that the networking opportunities at UCEM helped her get the “lay of the land,” so that she could more effectively participate in addressing academic racism at MIT.

“I am on the steering committee for the Strategic Action Plan that's being created around diversity, equity, and inclusion,” Lepe says. “A lot of the conversations and lived experiences that I've had [with] my fellow UCEM scholars has been really helpful in creating the priorities … under the official MIT strategic plan.”

Increasing representation in academia

In accordance with the Sloan Foundation objective to increase representation in academic faculty positions, Kolodziejski says that UCEM scholars attend the Institute on Teaching and Mentoring conference twice during their tenure at MIT. The conference is a three-day crash course in the fundamentals of being a professor, and covers topics ranging from mentoring students to writing proposals to presenting scholarship and giving presentations for employment.

Kolodziejski says that participation in this conference can be very clarifying for students, and for those who want to pursue an academic path, it helps outfit them with the tools they need upon graduating. However, Kolodziejski makes it clear that UCEM scholars are under no obligation to take an academic path.

For more information about UCEM, or if you are interested in becoming a UCEM scholar, please visit the UCEM website.

de MIT News https://ift.tt/3t7RgLo

The laws of physics and the physics of law

When Obiageli “Oby” Nwodoh arrived at MIT, she already felt at home. A native of Bedford, Massachusetts, she was the daughter of Thomas Nwodoh, a former MIT Media Lab researcher; her first physics teacher at Bedford High was an MIT alum, Joe Zahka; and she had participated in the Minority Introduction to Engineering and Science (MITES) program.

At MIT, she studied physics, excelling in research, data analytics, machine learning, and computer programming. “I fell in love with physics because it touched reality,” says Nwodoh. “I had a way of explaining the world in numbers when words were challenging. It was learning a new language and using it to describe the world.”

But her interests began to drift toward economic justice. Away from home, she slowly began to understand the economic inequality her family had always experienced. Though unaware as a child, she later learned her family benefited from certain antipoverty initiatives. “It helped us immensely with paying bills, funding extracurricular programs, and more,” she says.

The final click for her was during an internship with a defense contractor, which didn’t match with her political views. She wanted to take her career in a more people-focused direction, so as a sophomore, she enrolled in classes and extracurricular activities that stoked her interests in social justice, science activism, public policy, and equity and diversity.

That’s when dawned on this physics student that she wanted to be a lawyer. And she was surprised at how well the two disparate fields complemented each other.

“The law requires the critical thinking offered by physics," she says. "With both, there is always the need to observe global issues, obtain necessary data, and use some framework to find a solution. I wanted to solve hard world problems, but those that helped people. The law was an outlet to solving major world issues that I experienced as a child. I believe that in America, we are so comfortable with poverty. The law has been a way to change that, along with many other issues.”

Nwodoh worked for several summers with Greater Boston Legal Services’ low-income tax clinic, on cases pertaining to taxes, immigration, and employment. “It was meaningful because I was solving so many issues my own single mother faced,” she says.

By the second summer with GBLS, her work was helping with pandemic stimulus checks. “What really opened up my eyes was how the pandemic affected low-income populations,” she says. “The stimulus provided money for people, but I didn’t hear enough about people who didn’t receive the checks, including immigrants and many people receiving federal assistance through welfare. There were a lot of forgotten people in the pandemic. My work at GBLS solidified my interest in the law and how much impact it could have.” 

As a host for the Division of Student Life's podcast “MIT Is…” Nwodoh and her co-host Gabe Owens ’21 explored everything from MIT student life to global issues. She turned some of her research projects into podcasts about immigration, minority voter suppression, and the U.S. tax code, and another podcast turned into a research project where she examined how tax credits could be distributed in the state of New York to maximize payout. “I have dreams of starting my own show one day,” she says. 

Nwodoh later worked with the Harvard College Black Pre-Law Association, before helping launch the MIT Pre-Law Society to connect students with relevant career opportunities, classes, and resources. She also was active with the National Society of Black Engineers, and was a peer career advisor at MIT’s Career Advising and Professional Development office. “So many face imposter syndrome, both academically and professionally. Being able to hype a student up and reassure them of their capabilities always filled me with joy,” she says.

Her physics education continued to play a role in her legal work. When she researched policing and voting, and steered various projects as a virtual racial justice data analyst intern with the NAACP Legal Defense and Education Fund, she relied on her skills as a scientist.

“I saw how there was a plethora of data in the world, but not as many people who knew how to use it. Though my experience was short, it inspired me to learn more about data analytics and how it could be useful in the law, ethics, and other fields.”

After graduating this spring with a major in physics and a minor in political science, she became a program paralegal at Ropes and Gray in Chicago, and is looking into law schools. She hopes to focus on technology, such as the impact that algorithm bias has on vulnerable populations.

“I have cherished how being a physicist has prepared me to not be a physicist," she says. "Physics taught me the importance of problem-solving which could be applied in other areas of my life and interests. The technical skills could be used to ‘hack’ different parts of my world. Physics and the law come down to the same thing: interacting with the world in a profound way. MIT taught me that there is always space for my skills in every nook and cranny of the world’s biggest questions. I feel like my work as a physicist has prepared me to delve deeper into any issue, and holds me to an ethical standard of doing so.”

de MIT News https://ift.tt/38vmkLI

Using adversarial attacks to refine molecular energy predictions

Neural networks (NNs) are increasingly being used to predict new materials, the rate and yield of chemical reactions, and drug-target interactions, among others. For these applications, they are orders of magnitude faster than traditional methods such as quantum mechanical simulations. 

The price for this agility, however, is reliability. Because machine learning models only interpolate, they may fail when used outside the domain of training data.

But the part that worried Rafael Gómez-Bombarelli, the Jeffrey Cheah Career Development Professor in the MIT Department of Materials Science and Engineering, and graduate students Daniel Schwalbe-Koda and Aik Rui Tan was that establishing the limits of these machine learning (ML) models is tedious and labor-intensive. 

This is particularly true for predicting ‘‘potential energy surfaces” (PES), or the map of a molecule's energy in all its configurations. These surfaces encode the complexities of a molecule into flatlands, valleys, peaks, troughs, and ravines. The most stable configurations of a system are usually in the deep pits — quantum mechanical chasms from which atoms and molecules typically do not escape. 

In a recent Nature Communications paper, the research team presented a way to demarcate the “safe zone” of a neural network by using “adversarial attacks.” Adversarial attacks have been studied for other classes of problems, such as image classification, but this is the first time that they are being used to sample molecular geometries in a PES. 

“People have been using uncertainty for active learning for years in ML potentials. The key difference is that they need to run the full ML simulation and evaluate if the NN was reliable, and if it wasn't, acquire more data, retrain and re-simulate. Meaning that it takes a long time to nail down the right model, and one has to run the ML simulation many times” explains Gómez-Bombarelli.

The Gómez-Bombarelli lab at MIT works on a synergistic synthesis of first-principles simulation and machine learning that greatly speeds up this process. The actual simulations are run only for a small fraction of these molecules, and all those data are fed into a neural network that learns how to predict the same properties for the rest of the molecules. They have successfully demonstrated these methods for a growing class of novel materials that includes catalysts for producing hydrogen from water, cheaper polymer electrolytes for electric vehicles,  zeolites for molecular sieving, magnetic materials, and more. 

The challenge, however, is that these neural networks are only as smart as the data they are trained on.  Considering the PES map, 99 percent of the data may fall into one pit, totally missing valleys that are of more interest. 

Such wrong predictions can have disastrous consequences — think of a self-driving car that fails to identify a person crossing the street.

One way to find out the uncertainty of a model is to run the same data through multiple versions of it. 

For this project, the researchers had multiple neural networks predict the potential energy surface from the same data. Where the network is fairly sure of the prediction, the variation between the outputs of different networks is minimal and the surfaces largely converge. When the network is uncertain, the predictions of different models vary widely, producing a range of outputs, any of which could be the correct surface. 

The spread in the predictions of a “committee of neural networks” is the “uncertainty” at that point. A good model should not just indicate the best prediction, but also indicates the uncertainty about each of these predictions. It’s like the neural network says “this property for material A will have a value of X and I’m highly confident about it.”

This could have been an elegant solution but for the sheer scale of the combinatorial space. “Each simulation (which is ground feed for the neural network) may take from tens to thousands of CPU hours,” explains Schwalbe-Koda. For the results to be meaningful, multiple models must be run over a sufficient number of points in the PES, an extremely time-consuming process. 

Instead, the new approach only samples data points from regions of low prediction confidence, corresponding to specific geometries of a molecule. These molecules are then stretched or deformed slightly so that the uncertainty of the neural network committee is maximized. Additional data are computed for these molecules through simulations and then added to the initial training pool. 

The neural networks are trained again, and a new set of uncertainties are calculated. This process is repeated until the uncertainty associated with various points on the surface becomes well-defined and cannot be decreased any further. 

Gómez-Bombarelli explains, “We aspire to have a model that is perfect in the regions we care about (i.e., the ones that the simulation will visit) without having had to run the full ML simulation, by making sure that we make it very good in high-likelihood regions where it isn't.”

The paper presents several examples of this approach, including predicting complex supramolecular interactions in zeolites. These materials are cavernous crystals that act as molecular sieves with high shape selectivity. They find applications in catalysis, gas separation, and ion exchange, among others.

Because performing simulations of large zeolite structures is very costly, the researchers show how their method can provide significant savings in computational simulations. They used more than 15,000 examples to train a neural network to predict the potential energy surfaces for these systems. Despite the large cost required to generate the dataset, the final results are mediocre, with only around 80 percent of the neural network-based simulations being successful. To improve the performance of the model using traditional active learning methods, the researchers calculated an additional 5,000 data points, which improved the performance of the neural network potentials to 92 percent.

However, when the adversarial approach is used to retrain the neural networks, the authors saw a performance jump to 97 percent using only 500 extra points. That’s a remarkable result, the researchers say, especially considering that each of these extra points takes hundreds of CPU hours. 

This could be the most realistic method to probe the limits of models that researchers use to predict the behavior of materials and the progress of chemical reactions.

de MIT News https://ift.tt/3t4W1Wd

Study: Crowds can wise up to fake news

In the face of grave concerns about misinformation, social media networks and news organizations often employ fact-checkers to sort the real from the false. But fact-checkers can only assess a small portion of the stories floating around online.

A new study by MIT researchers suggests an alternate approach: Crowdsourced accuracy judgements from groups of normal readers can be virtually as effective as the work of professional fact-checkers.

“One problem with fact-checking is that there is just way too much content for professional fact-checkers to be able to cover, especially within a reasonable time frame,” says Jennifer Allen, a PhD student at the MIT Sloan School of Management  and co-author of a newly published paper detailing the study.

But the current study, examining over 200 news stories that Facebook’s algorithms had flagged for further scrutiny, may have found a way to address that problem, by using relatively small, politically balanced groups of lay readers to evaluate the headlines and lead sentences of news stories.

“We found it to be encouraging,” says Allen. “The average rating of a crowd of 10 to 15 people correlated as well with the fact-checkers’ judgments as the fact-checkers correlated with each other. This helps with the scalability problem because these raters were regular people without fact-checking training, and they just read the headlines and lead sentences without spending the time to do any research.”

That means the crowdsourcing method could be deployed widely — and cheaply. The study estimates that the cost of having readers evaluate news this way is about $0.90 per story.

“There’s no one thing that solves the problem of false news online,” says David Rand, a professor at MIT Sloan and senior co-author of the study. “But we’re working to add promising approaches to the anti-misinformation tool kit.”

The paper, “Scaling up Fact-Checking Using the Wisdom of Crowds,” is being published today in Science Advances. The co-authors are Allen; Antonio A. Arechar, a research scientist at the MIT Human Cooperation Lab; Gordon Pennycook, an assistant professor of behavioral science at University of Regina’s Hill/Levene Schools of Business; and Rand, who is the Erwin H. Schell Professor and a professor of management science and brain and cognitive sciences at MIT, and director of MIT’s Applied Cooperation Lab.

A critical mass of readers

To conduct the study, the researchers used 207 news articles that an internal Facebook algorithm identified as being in need of fact-checking, either because there was reason to believe they were problematic or simply because they were being widely shared or were about important topics like health. The experiment deployed 1,128 U.S. residents using Amazon’s Mechanical Turk platform.

Those participants were given the headline and lead sentence of 20 news stories and were asked seven questions — how much the story was “accurate,” “true,” “reliable,” “trustworthy,” “objective,” “unbiased,” and “describ[ing] an event that actually happened” — to generate an overall accuracy score about each news item.

At the same time, three professional fact-checkers were given all 207 stories —asked to evaluate the stories after researching them. In line with other studies on fact-checking, although the ratings of the fact-checkers were highly correlated with each other, their agreement was far from perfect. In about 49 percent of cases, all three fact-checkers agreed on the proper verdict about a story’s facticity; around 42 percent of the time, two of the three fact-checkers agreed; and about 9 percent of the time, the three fact-checkers each had different ratings.

Intriguingly, when the regular readers recruited for the study were sorted into groups with the same number of Democrats and Republicans, their average ratings were highly correlated with the professional fact-checkers’ ratings — and with at least a double-digit number of readers involved, the crowd’s ratings correlated as strongly with the fact-checkers as the fact-checkers’ did with each other.

“These readers weren’t trained in fact-checking, and they were only reading the headlines and lead sentences, and even so they were able to match the performance of the fact-checkers,” Allen says.

While it might seem initially surprising that a crowd of 12 to 20 readers could match the performance of professional fact-checkers, this is another example of a classic phenomenon: the wisdom of crowds. Across a wide range of applications, groups of laypeople have been found to match or exceed the performance of expert judgments. The current study shows this can occur even in the highly polarizing context of misinformation identification.

The experiment’s participants also took a political knowledge test and a test of their tendency to think analytically. Overall, the ratings of people who were better informed about civic issues and engaged in more analytical thinking were more closely aligned with the fact-checkers.

“People that engaged in more reasoning and were more knowledgeable agreed more with the fact-checkers,” Rand says. “And that was true regardless of whether they were Democrats or Republicans.”

Participation mechanisms

The scholars say the finding could be applied in many ways — and note that some social media behemoths are actively trying to make crowdsourcing work. Facebook has a program, called Community Review, where laypeople are hired to assess news content; Twitter has its own project, Birdwatch, soliciting reader input about the veracity of tweets. The wisdom of crowds can be used either to help apply public-facing labels to content, or to inform ranking algorithms and what content people are shown in the first place.

To be sure, the authors note, any organization using crowdsourcing needs to find a good mechanism for participation by readers. If participation is open to everyone, it is possible the crowdsourcing process could be unfairly influenced by partisans.

“We haven’t yet tested this in an environment where anyone can opt in,” Allen notes. “Platforms shouldn’t necessarily expect that other crowdsourcing strategies would produce equally positive results.”

On the other hand, Rand says, news and social media organizations would have to find ways to get a large enough groups of people actively evaluating news items, in order to make the crowdsourcing work.

“Most people don’t care about politics and care enough to try to influence things,” Rand says. “But the concern is that if you let people rate any content they want, then the only people doing it will be the ones who want to game the system. Still, to me, a bigger concern than being swamped by zealots is the problem that no one would do it. It is a classic public goods problem: Society at large benefits from people identifying misinformation, but why should users bother to invest the time and effort to give ratings?”

The study was supported, in part, by the William and Flora Hewlett Foundation, the John Templeton Foundation, and the Reset project of Omidyar Group’s Luminate Project Limited. Allen is a former Facebook employee who still has a financial interest in Facebook; other studies by Rand are supported, in part, by Google.

de MIT News https://ift.tt/3yAkfsy

“You’re surrounded by a community that cares about you”

As the sun broke through the clouds on a breezy Monday morning, first-year students and their families gathered on Kresge Oval for MIT’s Convocation, the Institute’s annual welcome to the incoming class.

The ceremony marked one of the first major events MIT has hosted on campus since the start of the Covid-19 pandemic. And while some aspects of the occasion were shaped by the ongoing pandemic — notably, masks were required of all who attended — the message to the 1,184 members of MIT’s Class of 2025 was one of hope, connection, and gratitude.

“Whether you know it or not, along with your suitcases, your boxes, your duffel bags, and your satchels, you also brought a gift to our community,” said President L. Rafael Reif in welcoming the incoming class. “You brought to us a gift of your talent, your energy, your curiosity, your creativity, and your drive. And you cannot imagine how grateful we are for that.”

As guests settled into their seats under a large and airy tent, the event opened with “Diary of a Pandemic Year,” a virtual performance that was written, composed, produced, and performed by hundreds of MIT musicians and community members.

“It is a homemade MIT masterpiece,” Reif said of the composition. “It offers a marvelous taste of so many things we love about MIT: a wonderful mix of people and backgrounds, the pleasure we take in making things together, and the energy and creative aspiration of everyone we meet.”

“Your new home”

Reif recalled first arriving at MIT in 1980 as an assistant professor of electrical engineering and computer science — “Which is course…?” he asked of the new students, to which they confidently shouted back, “Six!”

Having grown up in Caracas, Venezuela, with an accent that was shaped 2,000 miles south of Cambridge, Reif was anxious about fitting in at MIT. But he quickly found that, like him, many at MIT “came from somewhere else, and they cared about helping each other, and helping society.”

Joining Reif onstage were several senior members of the MIT administration: Provost Martin Schmidt, Chancellor Melissa Nobles, Vice Chancellor for Undergraduate and Graduate Education Ian Waitz, and Vice Chancellor and Dean for Student Life Suzy Nelson. Reif briefly introduced each of them, noting that they represent essential pieces of a rich support system available to MIT students.

“You’re surrounded by a community that cares about you,” he said. “All of us are dedicated to your success, and we believe in you.”

Moments to meander

Reif then introduced three members of the MIT faculty, who also happen to be MIT alumni: Shankar Raman ’86, section head and professor of literature; Evelyn Wang ’00, the Ford Professor of Engineering and head of the Department of Mechanical Engineering; and Steven D. Eppinger ’83, SM ’84, ScD ’88, the General Motors Leaders for Global Operations Professor of Management at MIT’s Sloan School of Management.

Raman, Wang, and Eppinger each spoke about living and learning at the Institute. For Raman, the MIT experience started out predictably enough. He recalled arriving as an undergraduate from India, “determined to major in Course 6 and emerge an electrical engineer.”

He also loved literature and philosophy, and on his way toward an engineering degree he sampled courses in German, poetry, and Western philosophy. After signing up for a filmmaking class, he stumbled upon MIT’s Department of Architecture, where the course was taught at the time. This encounter sprouted a new path, and Raman went on to earn degrees in both electrical engineering and architecture.

“Whatever your major, remember these four years are probably the only ones in your life where you can meander — where you can decide to not follow the main avenue, but to follow oblique paths and detours, to discover new areas of study,” he said.

His career continued to take unexpected turns. While pursuing a master’s in electrical engineering from the University of California at Berkeley, he realized that “my heart wasn’t fully in it.” So, he switched fields entirely, earning a PhD in literature from Stanford University. In 1995, he returned to MIT as a faculty member in the MIT Literature Section, and today serves as its head, teaching classes in Shakespeare, postcolonial fiction, and perspectives on artificial intelligence.

“I had come to MIT to become an electrical engineer, and I had certainly learned that,” Raman said. “But MIT also taught me how not to be one. And for that lesson, I will be forever grateful, and I hope it’s one you all will experience.”

“You’ve got this”

As a first-year herself, Evelyn Wang recalled setting out with energy, ready to “bring my ‘A’ game.” But her older brother, who also had attended MIT, warned her about “the wicked-hard problem sets,” and that she might not always get the A’s she was accustomed to in high school.

“Getting straight A’s is really, really tough,” Wang said. “You’re probably going to get a B, and maybe even a C or D, and that’s okay. I got an F on my first physics exam. Grades are only one way to measure what you’ll learn here.”

She offered tips for students to make the most of their time at MIT. The first is to be resilient and keep from dwelling on stress.

“Take breaks when you need to. Walk along Memorial Drive. Take a sailing class on the Charles. Tinker with a pet robot. Then get back to the problem sets,” Wang said. “You’ve got this.”

She also encouraged students to build a community — of friends, professors, and loved ones back home — who can support, advise, and ground them as they navigate the next four years.

Wang also reminded students to stay healthy, and pace themselves — advice she learned the hard way as an undergraduate. During a particularly grueling week, she recalled getting very little sleep while attempting to finish multiple class projects. She and her friends were fueled by cans of Mountain Dew, which they erected at the end of the ordeal, in a massive “victory tower.”

“Afterward, I slept for 36 hours straight,” Wang said. “Even when you are young, your body will fall apart if you do that every week. Please hydrate, and maybe drink less Mountain Dew than I did.”

“You are not alone”

As a newly arrived first-year at a similar MIT welcome event, Steven Eppinger remembered being given an obvious yet unsettling reality check.

“A speaker warned us, ‘half of you will be at the bottom half of the class,’” Eppinger said, drawing laughter from the crowd after a beat. “That statistical reality really struck me. Here we were, all these highly accomplished students, being told we may be average or worse. How could I process that?”

He did so by being open to imperfection. He came to MIT on a chemistry scholarship and had been the top chemistry student in both his high school and his state. At MIT, though, he quickly learned to redefine his expectations. “I was not devastated to score poorly on several chemistry exams in my first year,” he said.

Instead, he expanded his interests, by pledging a fraternity, joining the crew team, and participating in design challenges, a talent show, and even some campus hacks — all of which gave him a sense of community and helped to put his heavy courseload into perspective.

He encouraged the Class of 2025 to explore, and to reach out — to study groups, teaching assistants, advisors, and MIT’s Student Support Services — for help along the way.

“You are not alone in this journey,” said Eppinger, closing with a hopeful vision for the future:

“All of you are going to play a role in changing the world, through science and engineering, and a range of humanitarian endeavours,” he said. “You are going to be people of great consequence, who will do great things.”

de MIT News https://ift.tt/3kIISOx