National Science Foundation (NSF)

Subterms

    More stories

    • 63 Shares189 Views

      in Data Management & Statistics

      Hallucinating to better text translation

      by

      As babies, we babble and imitate our way to learning languages. We don’t start off reading raw text, which requires fundamental knowledge and understanding about the world, as well as the advanced ability to interpret and infer descriptions and relationships. Rather, humans begin our language journey slowly, by pointing and interacting with our environment, basing our words and perceiving their meaning through the context of the physical and social world. Eventually, we can craft full sentences to communicate complex ideas.

      Similarly, when humans begin learning and translating into another language, the incorporation of other sensory information, like multimedia, paired with the new and unfamiliar words, like flashcards with images, improves language acquisition and retention. Then, with enough practice, humans can accurately translate new, unseen sentences in context without the accompanying media; however, imagining a picture based on the original text helps.

      This is the basis of a new machine learning model, called VALHALLA, by researchers from MIT, IBM, and the University of California at San Diego, in which a trained neural network sees a source sentence in one language, hallucinates an image of what it looks like, and then uses both to translate into a target language. The team found that their method demonstrates improved accuracy of machine translation over text-only translation. Further, it provided an additional boost for cases with long sentences, under-resourced languages, and instances where part of the source sentence is inaccessible to the machine translator.

      As a core task within the AI field of natural language processing (NLP), machine translation is an “eminently practical technology that’s being used by millions of people every day,” says study co-author Yoon Kim, assistant professor in MIT’s Department of Electrical Engineering and Computer Science with affiliations in the Computer Science and Artificial Intelligence Laboratory (CSAIL) and the MIT-IBM Watson AI Lab. With recent, significant advances in deep learning, “there’s been an interesting development in how one might use non-text information — for example, images, audio, or other grounding information — to tackle practical tasks involving language” says Kim, because “when humans are performing language processing tasks, we’re doing so within a grounded, situated world.” The pairing of hallucinated images and text during inference, the team postulated, imitates that process, providing context for improved performance over current state-of-the-art techniques, which utilize text-only data.

      This research will be presented at the IEEE / CVF Computer Vision and Pattern Recognition Conference this month. Kim’s co-authors are UC San Diego graduate student Yi Li and Professor Nuno Vasconcelos, along with research staff members Rameswar Panda, Chun-fu “Richard” Chen, Rogerio Feris, and IBM Director David Cox of IBM Research and the MIT-IBM Watson AI Lab.

      Learning to hallucinate from images

      When we learn new languages and to translate, we’re often provided with examples and practice before venturing out on our own. The same is true for machine-translation systems; however, if images are used during training, these AI methods also require visual aids for testing, limiting their applicability, says Panda.

      “In real-world scenarios, you might not have an image with respect to the source sentence. So, our motivation was basically: Instead of using an external image during inference as input, can we use visual hallucination — the ability to imagine visual scenes — to improve machine translation systems?” says Panda.

      To do this, the team used an encoder-decoder architecture with two transformers, a type of neural network model that’s suited for sequence-dependent data, like language, that can pay attention key words and semantics of a sentence. One transformer generates a visual hallucination, and the other performs multimodal translation using outputs from the first transformer.

      During training, there are two streams of translation: a source sentence and a ground-truth image that is paired with it, and the same source sentence that is visually hallucinated to make a text-image pair. First the ground-truth image and sentence are tokenized into representations that can be handled by transformers; for the case of the sentence, each word is a token. The source sentence is tokenized again, but this time passed through the visual hallucination transformer, outputting a hallucination, a discrete image representation of the sentence. The researchers incorporated an autoregression that compares the ground-truth and hallucinated representations for congruency — e.g., homonyms: a reference to an animal “bat” isn’t hallucinated as a baseball bat. The hallucination transformer then uses the difference between them to optimize its predictions and visual output, making sure the context is consistent.

      The two sets of tokens are then simultaneously passed through the multimodal translation transformer, each containing the sentence representation and either the hallucinated or ground-truth image. The tokenized text translation outputs are compared with the goal of being similar to each other and to the target sentence in another language. Any differences are then relayed back to the translation transformer for further optimization.

      For testing, the ground-truth image stream drops off, since images likely wouldn’t be available in everyday scenarios.

      “To the best of our knowledge, we haven’t seen any work which actually uses a hallucination transformer jointly with a multimodal translation system to improve machine translation performance,” says Panda.

      Visualizing the target text

      To test their method, the team put VALHALLA up against other state-of-the-art multimodal and text-only translation methods. They used public benchmark datasets containing ground-truth images with source sentences, and a dataset for translating text-only news articles. The researchers measured its performance over 13 tasks, ranging from translation on well-resourced languages (like English, German, and French), under-resourced languages (like English to Romanian) and non-English (like Spanish to French). The group also tested varying transformer model sizes, how accuracy changes with the sentence length, and translation under limited textual context, where portions of the text were hidden from the machine translators.

      The team observed significant improvements over text-only translation methods, improving data efficiency, and that smaller models performed better than the larger base model. As sentences became longer, VALHALLA’s performance over other methods grew, which the researchers attributed to the addition of more ambiguous words. In cases where part of the sentence was masked, VALHALLA could recover and translate the original text, which the team found surprising.

      Further unexpected findings arose: “Where there weren’t as many training [image and] text pairs, [like for under-resourced languages], improvements were more significant, which indicates that grounding in images helps in low-data regimes,” says Kim. “Another thing that was quite surprising to me was this improved performance, even on types of text that aren’t necessarily easily connectable to images. For example, maybe it’s not so surprising if this helps in translating visually salient sentences, like the ‘there is a red car in front of the house.’ [However], even in text-only [news article] domains, the approach was able to improve upon text-only systems.”

      While VALHALLA performs well, the researchers note that it does have limitations, requiring pairs of sentences to be annotated with an image, which could make it more expensive to obtain. It also performs better in its ground domain and not the text-only news articles. Moreover, Kim and Panda note, a technique like VALHALLA is still a black box, with the assumption that hallucinated images are providing helpful information, and the team plans to investigate what and how the model is learning in order to validate their methods.

      In the future, the team plans to explore other means of improving translation. “Here, we only focus on images, but there are other types of a multimodal information — for example, speech, video or touch, or other sensory modalities,” says Panda. “We believe such multimodal grounding can lead to even more efficient machine translation models, potentially benefiting translation across many low-resource languages spoken in the world.”

      This research was supported, in part, by the MIT-IBM Watson AI Lab and the National Science Foundation. More

    • 125 Shares199 Views

      in Data Management & Statistics

      Making data visualization more accessible for blind and low-vision individuals

      by

      Data visualizations on the web are largely inaccessible for blind and low-vision individuals who use screen readers, an assistive technology that reads on-screen elements as text-to-speech. This excludes millions of people from the opportunity to probe and interpret insights that are often presented through charts, such as election results, health statistics, and economic indicators. 

      When a designer attempts to make a visualization accessible, best practices call for including a few sentences of text that describe the chart and a link to the underlying data table — a far cry from the rich reading experience available to sighted users.

      An interdisciplinary team of researchers from MIT and elsewhere is striving to create screen-reader-friendly data visualizations that offer a similarly rich experience. They prototyped several visualization structures that provide text descriptions at varying levels of detail, enabling a screen-reader user to drill down from high-level data to more detailed information using just a few keystrokes.

      The MIT team embarked on an iterative co-design process with collaborator Daniel Hajas, a researcher at University College London who works with the Global Disability Innovation Hub and lost his sight at age 16. They collaborated to develop prototypes and ran a detailed user study with blind and low-vision individuals to gather feedback.

      “Researchers might see some connections between problems and be aware of potential solutions, but very often they miss it by a little bit. Insights from people who have the lived experience of a certain specific, measurable problem are really important for a lot of disability-related solutions. I think we found a really nice fit,” says Hajas.

      They created a framework to help designers think systematically about how to develop accessible visualizations. In the future, they plan to use their prototypes and design framework to build a user-friendly tool that could convert visualizations into accessible formats.

      MIT collaborators include co-lead authors and Computer Science and Artificial Intelligence Laboratory (CSAIL) graduate students Jonathan Zong, Crystal Lee, and Alan Lundgard, as well as JiWoong Jang, an undergraduate at Carnegie Mellon University who worked on this project during MIT’s Summer Research Program (MSRP), and senior author Arvind Satyanarayan, assistant professor of computer science who leads the Visualization Group in CSAIL. The research paper, which will be presented at the Eurographics Conference on Visualization, won a best paper honorable mention award.

      “Push what is possible”

      The researchers defined three design dimensions as key to making accessible visualizations: structure, navigation, and description. Structure involves arranging the information into a hierarchy. Navigation refers to how the user moves through different levels of detail. Description is how the information is spoken, including how much information is conveyed.

      Using these design dimensions, they developed several visualization prototypes that emphasized ease-of-navigation for screen-reader users. One prototype, known as multiview, enabled individuals to use the up and down arrows to navigate between different levels of information (like the chart title as the top level, the legend as the second level, etc.), and the right and left arrow keys to cycle through information on the same level (such as adjacent scatterplots). Another prototype, known as target, included the same arrow key navigation but also a drop-down menu of key chart locations so the user could quickly jump to an area of interest.

      “Our goal is not just to work within existing standards to make them serviceable. We really set out to do grounded speculation and imagine where we can push what is possible with these existing standards. We didn’t want to limit ourselves to refitting tools that were designed for images,” says Zong.

      They tested these prototypes and an accessible data table, the existing best practice for accessible visualizations, with 13 blind and visually impaired screen-reader users. They asked users to rate each tool on several criteria, including how easy it was to learn and how easy it was to locate data or answer questions.

      “One thing I thought was really interesting was how much people were constantly testing their own hypotheses or trying to make specific patterns as they moved through the visualization. The implication for navigation is that you want to be able to orient yourself within the visualization so you know where the limits are,” says Lee. “Can you accurately and easily know where the walls are in the room you are exploring?”

      Improved insights

      Users said both prototypes enabled them to more rapidly identify patterns in the data. Scrolling from a high level to deeper levels of information helped them gain insights more easily than when browsing the data table, they said. They also enjoyed faster navigation using the menu in the target prototype.

      But the data table got top marks for ease of use.

      “I expected people to be disappointed with the everyday tools when compared to the new prototypes, but they still clung to the data table a bit, likely because of their familiarity with it. That shows that principles like familiarity, learnability, and usability still matter. No matter how ‘good’ our new invention is, if it is not easy enough to learn, people might stick with an older version,” Hajas says.

      Drawing on these insights, the researchers are refining the prototypes and using them to build a software package that can be used with existing design tools to give visualizations an accessible, navigable structure.

      They also want to explore multimodal solutions. Some study participants used different devices together, like screen readers and braille displays, or data sonification tools that convey information using non-speech audio. How these tools can complement each other when applied to a visualization is still an open question, Zong says.

      In the long-run, they hope their work might lead to careful rethinking of web accessibility standards.

      “There is no one-size-fits-all solution for accessibility. While existing standards don’t presume that, they only offer simple approaches, like data tables and alt text. One of the key benefits of our research contribution is that we are proposing a framework — different preferences and data representations are situated at different points in this design space,” says Lundgard.

      “We have been working hard toward reducing the inequities that screen-reader users face when extracting information from online data visualizations for the past few years. So, we are really appreciative of this work and the knowledge that it adds to the existing literature,” says Ather Sharif, a graduate student who researches accessibility and visualization in the labs of professors Jacob Wobbrock and Katharina Reinecke at the Paul G. Allen School of Computer Science and Engineering of the University of Washington at Seattle, and who was not involved with this work.

      “I like to think of it as a movement where we’re all finally coming together and improving the experiences of a demographic that has been largely ignored, especially when presenting data through visualizations. Kudos to Jonathan, Arvind, and their team for this insightful and timely work! I am looking forward to what’s next,” adds Sharif, who is lead author of several recent papers related to accessible data visualizations.

      Amy Bower, a senior scientist in the Department of Physical Oceanography at the Woods Hole Oceanographic Institution who suffers from a degenerative retinal disease and uses a screen reader extensively in her work as a researcher and also for basic living tasks, found the researchers’ explanations of the importance of co-design to be powerful and compelling.  

      “As a blind scientist, I’m constantly searching for effective tools that will allow me to access the information conveyed in data visualizations. The layered approach taken by these researchers, which provides the option to get the ‘big picture’ from the data as well as drill down into the data points themselves, allows the user to choose how they want to explore the data,” says Bower, who also was not involved with this work. “I think the ability to freely explore the data is necessary not just to learn the ‘story’ that the data are telling, but to allow a blind researcher such as myself to formulate the next questions that need to be tackled to advance understanding in any field of study.”

      This work was supported, in part, by the National Science Foundation.   More

    • 125 Shares189 Views

      in Data Management & Statistics

      Is it topological? A new materials database has the answer

      by

      What will it take to make our electronics smarter, faster, and more resilient? One idea is to build them from materials that are topological.

      Topology stems from a branch of mathematics that studies shapes that can be manipulated or deformed without losing certain core properties. A donut is a common example: If it were made of rubber, a donut could be twisted and squeezed into a completely new shape, such as a coffee mug, while retaining a key trait — namely, its center hole, which takes the form of the cup’s handle. The hole, in this case, is a topological trait, robust against certain deformations.

      In recent years, scientists have applied concepts of topology to the discovery of materials with similarly robust electronic properties. In 2007, researchers predicted the first electronic topological insulators — materials in which electrons that behave in ways that are “topologically protected,” or persistent in the face of certain disruptions.

      Since then, scientists have searched for more topological materials with the aim of building better, more robust electronic devices. Until recently, only a handful of such materials were identified, and were therefore assumed to be a rarity.

      Now researchers at MIT and elsewhere have discovered that, in fact, topological materials are everywhere, if you know how to look for them.

      In a paper published today in Science, the team, led by Nicolas Regnault of Princeton University and the École Normale Supérieure Paris, reports harnessing the power of multiple supercomputers to map the electronic structure of more than 96,000 natural and synthetic crystalline materials. They applied sophisticated filters to determine whether and what kind of topological traits exist in each structure.

      Overall, they found that 90 percent of all known crystalline structures contain at least one topological property, and more than 50 percent of all naturally occurring materials exhibit some sort of topological behavior.

      “We found there’s a ubiquity — topology is everywhere,” says Benjamin Wieder, the study’s co-lead, and a postdoc in MIT’s Department of Physics.

      The team has compiled the newly identified materials into a new, freely accessible Topological Materials Database resembling a periodic table of topology. With this new library, scientists can quickly search materials of interest for any topological properties they might hold, and harness them to build ultra-low-power transistors, new magnetic memory storage, and other devices with robust electronic properties.

      The paper includes co-lead author Maia Vergniory of the Donostia International Physics Center, Luis Elcoro of the University of Basque Country, Stuart Parkin and Claudia Felser of the Max Planck Institute, and Andrei Bernevig of Princeton University.

      Beyond intuition

      The new study was motivated by a desire to speed up the traditional search for topological materials.

      “The way the original materials were found was through chemical intuition,” Wieder says. “That approach had a lot of early successes. But as we theoretically predicted more kinds of topological phases, it seemed intuition wasn’t getting us very far.”

      Wieder and his colleagues instead utilized an efficient and systematic method to root out signs of topology, or robust electronic behavior, in all known crystalline structures, also known as inorganic solid-state materials.

      For their study, the researchers looked to the Inorganic Crystal Structure Database, or ICSD, a repository into which researchers enter the atomic and chemical structures of crystalline materials that they have studied. The database includes materials found in nature, as well as those that have been synthesized and manipulated in the lab. The ICSD is currently the largest materials database in the world, containing over 193,000 crystals whose structures have been mapped and characterized.

      The team downloaded the entire ICSD, and after performing some data cleaning to weed out structures with corrupted files or incomplete data, the researchers were left with just over 96,000 processable structures. For each of these structures, they performed a set of calculations based on fundamental knowledge of the relation between chemical constituents, to produce a map of the material’s electronic structure, also known as the electron band structure.

      The team was able to efficiently carry out the complicated calculations for each structure using multiple supercomputers, which they then employed to perform a second set of operations, this time to screen for various known topological phases, or persistent electrical behavior in each crystal material.

      “We’re looking for signatures in the electronic structure in which certain robust phenomena should occur in this material,” explains Wieder, whose previous work involved refining and expanding the screening technique, known as topological quantum chemistry.

      From their high-throughput analysis, the team quickly discovered a surprisingly large number of materials that are naturally topological, without any experimental manipulation, as well as materials that can be manipulated, for instance with light or chemical doping, to exhibit some sort of robust electronic behavior. They also discovered a handful of materials that contained more than one topological state when exposed to certain conditions.

      “Topological phases of matter in 3D solid-state materials have been proposed as venues for observing and manipulating exotic effects, including the interconversion of electrical current and electron spin, the tabletop simulation of exotic theories from high-energy physics, and even, under the right conditions, the storage and manipulation of quantum information,” Wieder notes. 

      For experimentalists who are studying such effects, Wieder says the team’s new database now reveals a menagerie of new materials to explore.

      This research was funded, in part, by the U.S. Department of Energy, the National Science Foundation, and the Office of Naval Research. More

    • 125 Shares129 Views

      in Data Management & Statistics

      System helps severely motor-impaired individuals type more quickly and accurately

      by

      In 1995, French fashion magazine editor Jean-Dominique Bauby suffered a seizure while driving a car, which left him with a condition known as locked-in syndrome, a neurological disease in which the patient is completely paralyzed and can only move muscles that control the eyes.

      Bauby, who had signed a book contract shortly before his accident, wrote the memoir “The Diving Bell and the Butterfly” using a dictation system in which his speech therapist recited the alphabet and he would blink when she said the correct letter. They wrote the 130-page book one blink at a time.

      Technology has come a long way since Bauby’s accident. Many individuals with severe motor impairments caused by locked-in syndrome, cerebral palsy, amyotrophic lateral sclerosis, or other conditions can communicate using computer interfaces where they select letters or words in an onscreen grid by activating a single switch, often by pressing a button, releasing a puff of air, or blinking.

      But these row-column scanning systems are very rigid, and, similar to the technique used by Bauby’s speech therapist, they highlight each option one at a time, making them frustratingly slow for some users. And they are not suitable for tasks where options can’t be arranged in a grid, like drawing, browsing the web, or gaming.

      A more flexible system being developed by researchers at MIT places individual selection indicators next to each option on a computer screen. The indicators can be placed anywhere — next to anything someone might click with a mouse — so a user does not need to cycle through a grid of choices to make selections. The system, called Nomon, incorporates probabilistic reasoning to learn how users make selections, and then adjusts the interface to improve their speed and accuracy.

      Participants in a user study were able to type faster using Nomon than with a row-column scanning system. The users also performed better on a picture selection task, demonstrating how Nomon could be used for more than typing.

      “It is so cool and exciting to be able to develop software that has the potential to really help people. Being able to find those signals and turn them into communication as we are used to it is a really interesting problem,” says senior author Tamara Broderick, an associate professor in the MIT 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.

      Joining Broderick on the paper are lead author Nicholas Bonaker, an EECS graduate student; Emli-Mari Nel, head of innovation and machine learning at Averly and a visiting lecturer at the University of Witwatersrand in South Africa; and Keith Vertanen, an associate professor at Michigan Tech. The research is being presented at the ACM Conference on Human Factors in Computing Systems.

      On the clock

      In the Nomon interface, a small analog clock is placed next to every option the user can select. (A gnomon is the part of a sundial that casts a shadow.) The user looks at one option and then clicks their switch when that clock’s hand passes a red “noon” line. After each click, the system changes the phases of the clocks to separate the most probable next targets. The user clicks repeatedly until their target is selected.

      When used as a keyboard, Nomon’s machine-learning algorithms try to guess the next word based on previous words and each new letter as the user makes selections.

      Broderick developed a simplified version of Nomon several years ago but decided to revisit it to make the system easier for motor-impaired individuals to use. She enlisted the help of then-undergraduate Bonaker to redesign the interface.

      They first consulted nonprofit organizations that work with motor-impaired individuals, as well as a motor-impaired switch user, to gather feedback on the Nomon design.

      Then they designed a user study that would better represent the abilities of motor-impaired individuals. They wanted to make sure to thoroughly vet the system before using much of the valuable time of motor-impaired users, so they first tested on non-switch users, Broderick explains.

      Switching up the switch

      To gather more representative data, Bonaker devised a webcam-based switch that was harder to use than simply clicking a key. The non-switch users had to lean their bodies to one side of the screen and then back to the other side to register a click.

      “And they have to do this at precisely the right time, so it really slows them down. We did some empirical studies which showed that they were much closer to the response times of motor-impaired individuals,” Broderick says.

      They ran a 10-session user study with 13 non-switch participants and one single-switch user with an advanced form of spinal muscular dystrophy. In the first nine sessions, participants used Nomon and a row-column scanning interface for 20 minutes each to perform text entry, and in the 10th session they used the two systems for a picture selection task.

      Non-switch users typed 15 percent faster using Nomon, while the motor-impaired user typed even faster than the non-switch users. When typing unfamiliar words, the users were 20 percent faster overall and made half as many errors. In their final session, they were able to complete the picture selection task 36 percent faster using Nomon.

      “Nomon is much more forgiving than row-column scanning. With row-column scanning, even if you are just slightly off, now you’ve chosen B instead of A and that’s an error,” Broderick says.

      Adapting to noisy clicks

      With its probabilistic reasoning, Nomon incorporates everything it knows about where a user is likely to click to make the process faster, easier, and less error-prone. For instance, if the user selects “Q,” Nomon will make it as easy as possible for the user to select “U” next.

      Nomon also learns how a user clicks. So, if the user always clicks a little after the clock’s hand strikes noon, the system adapts to that in real time. It also adapts to noisiness. If a user’s click is often off the mark, the system requires extra clicks to ensure accuracy.

      This probabilistic reasoning makes Nomon powerful but also requires a higher click-load than row-column scanning systems. Clicking multiple times can be a trying task for severely motor-impaired users.

      Broderick hopes to reduce the click-load by incorporating gaze tracking into Nomon, which would give the system more robust information about what a user might choose next based on which part of the screen they are looking at. The researchers also want to find a better way to automatically adjust the clock speeds to help users be more accurate and efficient.

      They are working on a new series of studies in which they plan to partner with more motor-impaired users.

      “So far, the feedback from motor-impaired users has been invaluable to us; we’re very grateful to the motor-impaired user who commented on our initial interface and the separate motor-impaired user who participated in our study. We’re currently extending our study to work with a bigger and more diverse group of our target population. With their help, we’re already making further improvements to our interface and working to better understand the performance of Nomon,” she says.

      “Nonspeaking individuals with motor disabilities are currently not provided with efficient communication solutions for interacting with either speaking partners or computer systems. This ‘communication gap’ is a known unresolved problem in human-computer interaction, and so far there are no good solutions. This paper demonstrates that a highly creative approach underpinned by a statistical model can provide tangible performance gains to the users who need it the most: nonspeaking individuals reliant on a single switch to communicate,” says Per Ola Kristensson, professor of interactive systems engineering at Cambridge University, who was not involved with this research. “The paper also demonstrates the value of complementing insights from computational experiments with the involvement of end-users and other stakeholders in the design process. I find this a highly creative and important paper in an area where it is notoriously difficult to make significant progress.”

      This research was supported, in part, by the Seth Teller Memorial Fund to Advanced Technology for People with Disabilities, a Peter J. Eloranta Summer Undergraduate Research Fellowship, the MIT Quest for Intelligence, and the National Science Foundation. More

    • 88 Shares189 Views

      in Data Management & Statistics

      Computational modeling guides development of new materials

      by

      Metal-organic frameworks, a class of materials with porous molecular structures, have a variety of possible applications, such as capturing harmful gases and catalyzing chemical reactions. Made of metal atoms linked by organic molecules, they can be configured in hundreds of thousands of different ways.

      To help researchers sift through all of the possible metal-organic framework (MOF) structures and help identify the ones that would be most practical for a particular application, a team of MIT computational chemists has developed a model that can analyze the features of a MOF structure and predict if it will be stable enough to be useful.

      The researchers hope that these computational predictions will help cut the development time of new MOFs.

      “This will allow researchers to test the promise of specific materials before they go through the trouble of synthesizing them,” says Heather Kulik, an associate professor of chemical engineering at MIT.

      The MIT team is now working to develop MOFs that could be used to capture methane gas and convert it to useful compounds such as fuels.

      The researchers described their new model in two papers, one in the Journal of the American Chemical Society and one in Scientific Data. Graduate students Aditya Nandy and Gianmarco Terrones are the lead authors of the Scientific Data paper, and Nandy is also the lead author of the JACS paper. Kulik is the senior author of both papers.

      Modeling structure

      MOFs consist of metal atoms joined by organic molecules called linkers to create a rigid, cage-like structure. The materials also have many pores, which makes them useful for catalyzing reactions involving gases but can also make them less structurally stable.

      “The limitation in seeing MOFs realized at industrial scale is that although we can control their properties by controlling where each atom is in the structure, they’re not necessarily that stable, as far as materials go,” Kulik says. “They’re very porous and they can degrade under realistic conditions that we need for catalysis.”

      Scientists have been working on designing MOFs for more than 20 years, and thousands of possible structures have been published. A centralized repository contains about 10,000 of these structures but is not linked to any of the published findings on the properties of those structures.

      Kulik, who specializes in using computational modeling to discover structure-property relationships of materials, wanted to take a more systematic approach to analyzing and classifying the properties of MOFs.

      “When people make these now, it’s mostly trial and error. The MOF dataset is really promising because there are so many people excited about MOFs, so there’s so much to learn from what everyone’s been working on, but at the same time, it’s very noisy and it’s not systematic the way it’s reported,” she says.

      Kulik and her colleagues set out to analyze published reports of MOF structures and properties using a natural-language-processing algorithm. Using this algorithm, they scoured nearly 4,000 published papers, extracting information on the temperature at which a given MOF would break down. They also pulled out data on whether particular MOFs can withstand the conditions needed to remove solvents used to synthesize them and make sure they become porous.

      Once the researchers had this information, they used it to train two neural networks to predict MOFs’ thermal stability and stability during solvent removal, based on the molecules’ structure.

      “Before you start working with a material and thinking about scaling it up for different applications, you want to know will it hold up, or is it going to degrade in the conditions I would want to use it in?” Kulik says. “Our goal was to get better at predicting what makes a stable MOF.”

      Better stability

      Using the model, the researchers were able to identify certain features that influence stability. In general, simpler linkers with fewer chemical groups attached to them are more stable. Pore size is also important: Before the researchers did their analysis, it had been thought that MOFs with larger pores might be too unstable. However, the MIT team found that large-pore MOFs can be stable if other aspects of their structure counteract the large pore size.

      “Since MOFs have so many things that can vary at the same time, such as the metal, the linkers, the connectivity, and the pore size, it is difficult to nail down what governs stability across different families of MOFs,” Nandy says. “Our models enable researchers to make predictions on existing or new materials, many of which have yet to be made.”

      The researchers have made their data and models available online. Scientists interested in using the models can get recommendations for strategies to make an existing MOF more stable, and they can also add their own data and feedback on the predictions of the models.

      The MIT team is now using the model to try to identify MOFs that could be used to catalyze the conversion of methane gas to methanol, which could be used as fuel. Kulik also plans to use the model to create a new dataset of hypothetical MOFs that haven’t been built before but are predicted to have high stability. Researchers could then screen this dataset for a variety of properties.

      “People are interested in MOFs for things like quantum sensing and quantum computing, all sorts of different applications where you need metals distributed in this atomically precise way,” Kulik says.

      The research was funded by DARPA, the U.S. Office of Naval Research, the U.S. Department of Energy, a National Science Foundation Graduate Research Fellowship, a Career Award at the Scientific Interface from the Burroughs Wellcome Fund, and an AAAS Marion Milligan Mason Award. More

    • 75 Shares149 Views

      in Data Management & Statistics

      Can machine-learning models overcome biased datasets?

      by

      Artificial intelligence systems may be able to complete tasks quickly, but that doesn’t mean they always do so fairly. If the datasets used to train machine-learning models contain biased data, it is likely the system could exhibit that same bias when it makes decisions in practice.

      For instance, if a dataset contains mostly images of white men, then a facial-recognition model trained with these data may be less accurate for women or people with different skin tones.

      A group of researchers at MIT, in collaboration with researchers at Harvard University and Fujitsu Ltd., sought to understand when and how a machine-learning model is capable of overcoming this kind of dataset bias. They used an approach from neuroscience to study how training data affects whether an artificial neural network can learn to recognize objects it has not seen before. A neural network is a machine-learning model that mimics the human brain in the way it contains layers of interconnected nodes, or “neurons,” that process data.

      The new results show that diversity in training data has a major influence on whether a neural network is able to overcome bias, but at the same time dataset diversity can degrade the network’s performance. They also show that how a neural network is trained, and the specific types of neurons that emerge during the training process, can play a major role in whether it is able to overcome a biased dataset.

      “A neural network can overcome dataset bias, which is encouraging. But the main takeaway here is that we need to take into account data diversity. We need to stop thinking that if you just collect a ton of raw data, that is going to get you somewhere. We need to be very careful about how we design datasets in the first place,” says Xavier Boix, a research scientist in the Department of Brain and Cognitive Sciences (BCS) and the Center for Brains, Minds, and Machines (CBMM), and senior author of the paper.  

      Co-authors include former MIT graduate students Timothy Henry, Jamell Dozier, Helen Ho, Nishchal Bhandari, and Spandan Madan, a corresponding author who is currently pursuing a PhD at Harvard; Tomotake Sasaki, a former visiting scientist now a senior researcher at Fujitsu Research; Frédo Durand, a professor of electrical engineering and computer science at MIT and a member of the Computer Science and Artificial Intelligence Laboratory; and Hanspeter Pfister, the An Wang Professor of Computer Science at the Harvard School of Enginering and Applied Sciences. The research appears today in Nature Machine Intelligence.

      Thinking like a neuroscientist

      Boix and his colleagues approached the problem of dataset bias by thinking like neuroscientists. In neuroscience, Boix explains, it is common to use controlled datasets in experiments, meaning a dataset in which the researchers know as much as possible about the information it contains.

      The team built datasets that contained images of different objects in varied poses, and carefully controlled the combinations so some datasets had more diversity than others. In this case, a dataset had less diversity if it contains more images that show objects from only one viewpoint. A more diverse dataset had more images showing objects from multiple viewpoints. Each dataset contained the same number of images.

      The researchers used these carefully constructed datasets to train a neural network for image classification, and then studied how well it was able to identify objects from viewpoints the network did not see during training (known as an out-of-distribution combination). 

      For example, if researchers are training a model to classify cars in images, they want the model to learn what different cars look like. But if every Ford Thunderbird in the training dataset is shown from the front, when the trained model is given an image of a Ford Thunderbird shot from the side, it may misclassify it, even if it was trained on millions of car photos.

      The researchers found that if the dataset is more diverse — if more images show objects from different viewpoints — the network is better able to generalize to new images or viewpoints. Data diversity is key to overcoming bias, Boix says.

      “But it is not like more data diversity is always better; there is a tension here. When the neural network gets better at recognizing new things it hasn’t seen, then it will become harder for it to recognize things it has already seen,” he says.

      Testing training methods

      The researchers also studied methods for training the neural network.

      In machine learning, it is common to train a network to perform multiple tasks at the same time. The idea is that if a relationship exists between the tasks, the network will learn to perform each one better if it learns them together.

      But the researchers found the opposite to be true — a model trained separately for each task was able to overcome bias far better than a model trained for both tasks together.

      “The results were really striking. In fact, the first time we did this experiment, we thought it was a bug. It took us several weeks to realize it was a real result because it was so unexpected,” he says.

      They dove deeper inside the neural networks to understand why this occurs.

      They found that neuron specialization seems to play a major role. When the neural network is trained to recognize objects in images, it appears that two types of neurons emerge — one that specializes in recognizing the object category and another that specializes in recognizing the viewpoint.

      When the network is trained to perform tasks separately, those specialized neurons are more prominent, Boix explains. But if a network is trained to do both tasks simultaneously, some neurons become diluted and don’t specialize for one task. These unspecialized neurons are more likely to get confused, he says.

      “But the next question now is, how did these neurons get there? You train the neural network and they emerge from the learning process. No one told the network to include these types of neurons in its architecture. That is the fascinating thing,” he says.

      That is one area the researchers hope to explore with future work. They want to see if they can force a neural network to develop neurons with this specialization. They also want to apply their approach to more complex tasks, such as objects with complicated textures or varied illuminations.

      Boix is encouraged that a neural network can learn to overcome bias, and he is hopeful their work can inspire others to be more thoughtful about the datasets they are using in AI applications.

      This work was supported, in part, by the National Science Foundation, a Google Faculty Research Award, the Toyota Research Institute, the Center for Brains, Minds, and Machines, Fujitsu Research, and the MIT-Sensetime Alliance on Artificial Intelligence. More

    • 75 Shares99 Views

      in Data Management & Statistics

      Avoiding shortcut solutions in artificial intelligence

      by

      If your Uber driver takes a shortcut, you might get to your destination faster. But if a machine learning model takes a shortcut, it might fail in unexpected ways.

      In machine learning, a shortcut solution occurs when the model relies on a simple characteristic of a dataset to make a decision, rather than learning the true essence of the data, which can lead to inaccurate predictions. For example, a model might learn to identify images of cows by focusing on the green grass that appears in the photos, rather than the more complex shapes and patterns of the cows.  

      A new study by researchers at MIT explores the problem of shortcuts in a popular machine-learning method and proposes a solution that can prevent shortcuts by forcing the model to use more data in its decision-making.

      By removing the simpler characteristics the model is focusing on, the researchers force it to focus on more complex features of the data that it hadn’t been considering. Then, by asking the model to solve the same task two ways — once using those simpler features, and then also using the complex features it has now learned to identify — they reduce the tendency for shortcut solutions and boost the performance of the model.

      One potential application of this work is to enhance the effectiveness of machine learning models that are used to identify disease in medical images. Shortcut solutions in this context could lead to false diagnoses and have dangerous implications for patients.

      “It is still difficult to tell why deep networks make the decisions that they do, and in particular, which parts of the data these networks choose to focus upon when making a decision. If we can understand how shortcuts work in further detail, we can go even farther to answer some of the fundamental but very practical questions that are really important to people who are trying to deploy these networks,” says Joshua Robinson, a PhD student in the Computer Science and Artificial Intelligence Laboratory (CSAIL) and lead author of the paper.

      Robinson wrote the paper with his advisors, senior author Suvrit Sra, the Esther and Harold E. Edgerton Career Development Associate Professor in the Department of Electrical Engineering and Computer Science (EECS) and a core member of the Institute for Data, Systems, and Society (IDSS) and the Laboratory for Information and Decision Systems; and Stefanie Jegelka, the X-Consortium Career Development Associate Professor in EECS and a member of CSAIL and IDSS; as well as University of Pittsburgh assistant professor Kayhan Batmanghelich and PhD students Li Sun and Ke Yu. The research will be presented at the Conference on Neural Information Processing Systems in December. 

      The long road to understanding shortcuts

      The researchers focused their study on contrastive learning, which is a powerful form of self-supervised machine learning. In self-supervised machine learning, a model is trained using raw data that do not have label descriptions from humans. It can therefore be used successfully for a larger variety of data.

      A self-supervised learning model learns useful representations of data, which are used as inputs for different tasks, like image classification. But if the model takes shortcuts and fails to capture important information, these tasks won’t be able to use that information either.

      For example, if a self-supervised learning model is trained to classify pneumonia in X-rays from a number of hospitals, but it learns to make predictions based on a tag that identifies the hospital the scan came from (because some hospitals have more pneumonia cases than others), the model won’t perform well when it is given data from a new hospital.     

      For contrastive learning models, an encoder algorithm is trained to discriminate between pairs of similar inputs and pairs of dissimilar inputs. This process encodes rich and complex data, like images, in a way that the contrastive learning model can interpret.

      The researchers tested contrastive learning encoders with a series of images and found that, during this training procedure, they also fall prey to shortcut solutions. The encoders tend to focus on the simplest features of an image to decide which pairs of inputs are similar and which are dissimilar. Ideally, the encoder should focus on all the useful characteristics of the data when making a decision, Jegelka says.

      So, the team made it harder to tell the difference between the similar and dissimilar pairs, and found that this changes which features the encoder will look at to make a decision.

      “If you make the task of discriminating between similar and dissimilar items harder and harder, then your system is forced to learn more meaningful information in the data, because without learning that it cannot solve the task,” she says.

      But increasing this difficulty resulted in a tradeoff — the encoder got better at focusing on some features of the data but became worse at focusing on others. It almost seemed to forget the simpler features, Robinson says.

      To avoid this tradeoff, the researchers asked the encoder to discriminate between the pairs the same way it had originally, using the simpler features, and also after the researchers removed the information it had already learned. Solving the task both ways simultaneously caused the encoder to improve across all features.

      Their method, called implicit feature modification, adaptively modifies samples to remove the simpler features the encoder is using to discriminate between the pairs. The technique does not rely on human input, which is important because real-world data sets can have hundreds of different features that could combine in complex ways, Sra explains.

      From cars to COPD

      The researchers ran one test of this method using images of vehicles. They used implicit feature modification to adjust the color, orientation, and vehicle type to make it harder for the encoder to discriminate between similar and dissimilar pairs of images. The encoder improved its accuracy across all three features — texture, shape, and color — simultaneously.

      To see if the method would stand up to more complex data, the researchers also tested it with samples from a medical image database of chronic obstructive pulmonary disease (COPD). Again, the method led to simultaneous improvements across all features they evaluated.

      While this work takes some important steps forward in understanding the causes of shortcut solutions and working to solve them, the researchers say that continuing to refine these methods and applying them to other types of self-supervised learning will be key to future advancements.

      “This ties into some of the biggest questions about deep learning systems, like ‘Why do they fail?’ and ‘Can we know in advance the situations where your model will fail?’ There is still a lot farther to go if you want to understand shortcut learning in its full generality,” Robinson says.

      This research is supported by the National Science Foundation, National Institutes of Health, and the Pennsylvania Department of Health’s SAP SE Commonwealth Universal Research Enhancement (CURE) program. More

    • 50 Shares149 Views

      in Data Management & Statistics

      Taming the data deluge

      by

      An oncoming tsunami of data threatens to overwhelm huge data-rich research projects on such areas that range from the tiny neutrino to an exploding supernova, as well as the mysteries deep within the brain. 

      When LIGO picks up a gravitational-wave signal from a distant collision of black holes and neutron stars, a clock starts ticking for capturing the earliest possible light that may accompany them: time is of the essence in this race. Data collected from electrical sensors monitoring brain activity are outpacing computing capacity. Information from the Large Hadron Collider (LHC)’s smashed particle beams will soon exceed 1 petabit per second. 

      To tackle this approaching data bottleneck in real-time, a team of researchers from nine institutions led by the University of Washington, including MIT, has received $15 million in funding to establish the Accelerated AI Algorithms for Data-Driven Discovery (A3D3) Institute. From MIT, the research team includes Philip Harris, assistant professor of physics, who will serve as the deputy director of the A3D3 Institute; Song Han, assistant professor of electrical engineering and computer science, who will serve as the A3D3’s co-PI; and Erik Katsavounidis, senior research scientist with the MIT Kavli Institute for Astrophysics and Space Research.

      Infused with this five-year Harnessing the Data Revolution Big Idea grant, and jointly funded by the Office of Advanced Cyberinfrastructure, A3D3 will focus on three data-rich fields: multi-messenger astrophysics, high-energy particle physics, and brain imaging neuroscience. By enriching AI algorithms with new processors, A3D3 seeks to speed up AI algorithms for solving fundamental problems in collider physics, neutrino physics, astronomy, gravitational-wave physics, computer science, and neuroscience. 

      “I am very excited about the new Institute’s opportunities for research in nuclear and particle physics,” says Laboratory for Nuclear Science Director Boleslaw Wyslouch. “Modern particle detectors produce an enormous amount of data, and we are looking for extraordinarily rare signatures. The application of extremely fast processors to sift through these mountains of data will make a huge difference in what we will measure and discover.”

      The seeds of A3D3 were planted in 2017, when Harris and his colleagues at Fermilab and CERN decided to integrate real-time AI algorithms to process the incredible rates of data at the LHC. Through email correspondence with Han, Harris’ team built a compiler, HLS4ML, that could run an AI algorithm in nanoseconds.

      “Before the development of HLS4ML, the fastest processing that we knew of was roughly a millisecond per AI inference, maybe a little faster,” says Harris. “We realized all the AI algorithms were designed to solve much slower problems, such as image and voice recognition. To get to nanosecond inference timescales, we recognized we could make smaller algorithms and rely on custom implementations with Field Programmable Gate Array (FPGA) processors in an approach that was largely different from what others were doing.”

      A few months later, Harris presented their research at a physics faculty meeting, where Katsavounidis became intrigued. Over coffee in Building 7, they discussed combining Harris’ FPGA with Katsavounidis’s use of machine learning for finding gravitational waves. FPGAs and other new processor types, such as graphics processing units (GPUs), accelerate AI algorithms to more quickly analyze huge amounts of data.

      “I had worked with the first FPGAs that were out in the market in the early ’90s and have witnessed first-hand how they revolutionized front-end electronics and data acquisition in big high-energy physics experiments I was working on back then,” recalls Katsavounidis. “The ability to have them crunch gravitational-wave data has been in the back of my mind since joining LIGO over 20 years ago.”

      Two years ago they received their first grant, and the University of Washington’s Shih-Chieh Hsu joined in. The team initiated the Fast Machine Lab, published about 40 papers on the subject, built the group to about 50 researchers, and “launched a whole industry of how to explore a region of AI that has not been explored in the past,” says Harris. “We basically started this without any funding. We’ve been getting small grants for various projects over the years. A3D3 represents our first large grant to support this effort.”  

      “What makes A3D3 so special and suited to MIT is its exploration of a technical frontier, where AI is implemented not in high-level software, but rather in lower-level firmware, reconfiguring individual gates to address the scientific question at hand,” says Rob Simcoe, director of MIT Kavli Institute for Astrophysics and Space Research and the Francis Friedman Professor of Physics. “We are in an era where experiments generate torrents of data. The acceleration gained from tailoring reprogrammable, bespoke computers at the processor level can advance real-time analysis of these data to new levels of speed and sophistication.”

      The Huge Data from the Large Hadron Collider 

      With data rates already exceeding 500 terabits per second, the LHC processes more data than any other scientific instrument on earth. Its future aggregate data rates will soon exceed 1 petabit per second, the biggest data rate in the world. 

      “Through the use of AI, A3D3 aims to perform advanced analyses, such as anomaly detection, and particle reconstruction on all collisions happening 40 million times per second,” says Harris.

      The goal is to find within all of this data a way to identify the few collisions out of the 3.2 billion collisions per second that could reveal new forces, explain how dark matter is formed, and complete the picture of how fundamental forces interact with matter. Processing all of this information requires a customized computing system capable of interpreting the collider information within ultra-low latencies.  

      “The challenge of running this on all of the 100s of terabits per second in real-time is daunting and requires a complete overhaul of how we design and implement AI algorithms,” says Harris. “With large increases in the detector resolution leading to data rates that are even larger the challenge of finding the one collision, among many, will become even more daunting.” 

      The Brain and the Universe

      Thanks to advances in techniques such as medical imaging and electrical recordings from implanted electrodes, neuroscience is also gathering larger amounts of data on how the brain’s neural networks process responses to stimuli and perform motor information. A3D3 plans to develop and implement high-throughput and low-latency AI algorithms to process, organize, and analyze massive neural datasets in real time, to probe brain function in order to enable new experiments and therapies.   

      With Multi-Messenger Astrophysics (MMA), A3D3 aims to quickly identify astronomical events by efficiently processing data from gravitational waves, gamma-ray bursts, and neutrinos picked up by telescopes and detectors. 

      The A3D3 researchers also include a multi-disciplinary group of 15 other researchers, including project lead the University of Washington, along with Caltech, Duke University, Purdue University, UC San Diego, University of Illinois Urbana-Champaign, University of Minnesota, and the University of Wisconsin-Madison. It will include neutrinos research at Icecube and DUNE, and visible astronomy at Zwicky Transient Facility, and will organize deep-learning workshops and boot camps to train students and researchers on how to contribute to the framework and widen the use of fast AI strategies.

      “We have reached a point where detector network growth will be transformative, both in terms of event rates and in terms of astrophysical reach and ultimately, discoveries,” says Katsavounidis. “‘Fast’ and ‘efficient’ is the only way to fight the ‘faint’ and ‘fuzzy’ that is out there in the universe, and the path for getting the most out of our detectors. A3D3 on one hand is going to bring production-scale AI to gravitational-wave physics and multi-messenger astronomy; but on the other hand, we aspire to go beyond our immediate domains and become the go-to place across the country for applications of accelerated AI to data-driven disciplines.” More