Computer science and technology

Subterms

    More stories

    • 125 Shares169 Views

      in Data Management & Statistics

      Multiple AI models help robots execute complex plans more transparently

      by

      Your daily to-do list is likely pretty straightforward: wash the dishes, buy groceries, and other minutiae. It’s unlikely you wrote out “pick up the first dirty dish,” or “wash that plate with a sponge,” because each of these miniature steps within the chore feels intuitive. While we can routinely complete each step without much thought, a robot requires a complex plan that involves more detailed outlines.

      MIT’s Improbable AI Lab, a group within the Computer Science and Artificial Intelligence Laboratory (CSAIL), has offered these machines a helping hand with a new multimodal framework: Compositional Foundation Models for Hierarchical Planning (HiP), which develops detailed, feasible plans with the expertise of three different foundation models. Like OpenAI’s GPT-4, the foundation model that ChatGPT and Bing Chat were built upon, these foundation models are trained on massive quantities of data for applications like generating images, translating text, and robotics.Unlike RT2 and other multimodal models that are trained on paired vision, language, and action data, HiP uses three different foundation models each trained on different data modalities. Each foundation model captures a different part of the decision-making process and then works together when it’s time to make decisions. HiP removes the need for access to paired vision, language, and action data, which is difficult to obtain. HiP also makes the reasoning process more transparent.

      What’s considered a daily chore for a human can be a robot’s “long-horizon goal” — an overarching objective that involves completing many smaller steps first — requiring sufficient data to plan, understand, and execute objectives. While computer vision researchers have attempted to build monolithic foundation models for this problem, pairing language, visual, and action data is expensive. Instead, HiP represents a different, multimodal recipe: a trio that cheaply incorporates linguistic, physical, and environmental intelligence into a robot.

      “Foundation models do not have to be monolithic,” says NVIDIA AI researcher Jim Fan, who was not involved in the paper. “This work decomposes the complex task of embodied agent planning into three constituent models: a language reasoner, a visual world model, and an action planner. It makes a difficult decision-making problem more tractable and transparent.”The team believes that their system could help these machines accomplish household chores, such as putting away a book or placing a bowl in the dishwasher. Additionally, HiP could assist with multistep construction and manufacturing tasks, like stacking and placing different materials in specific sequences.Evaluating HiP

      The CSAIL team tested HiP’s acuity on three manipulation tasks, outperforming comparable frameworks. The system reasoned by developing intelligent plans that adapt to new information.

      First, the researchers requested that it stack different-colored blocks on each other and then place others nearby. The catch: Some of the correct colors weren’t present, so the robot had to place white blocks in a color bowl to paint them. HiP often adjusted to these changes accurately, especially compared to state-of-the-art task planning systems like Transformer BC and Action Diffuser, by adjusting its plans to stack and place each square as needed.

      Another test: arranging objects such as candy and a hammer in a brown box while ignoring other items. Some of the objects it needed to move were dirty, so HiP adjusted its plans to place them in a cleaning box, and then into the brown container. In a third demonstration, the bot was able to ignore unnecessary objects to complete kitchen sub-goals such as opening a microwave, clearing a kettle out of the way, and turning on a light. Some of the prompted steps had already been completed, so the robot adapted by skipping those directions.

      A three-pronged hierarchy

      HiP’s three-pronged planning process operates as a hierarchy, with the ability to pre-train each of its components on different sets of data, including information outside of robotics. At the bottom of that order is a large language model (LLM), which starts to ideate by capturing all the symbolic information needed and developing an abstract task plan. Applying the common sense knowledge it finds on the internet, the model breaks its objective into sub-goals. For example, “making a cup of tea” turns into “filling a pot with water,” “boiling the pot,” and the subsequent actions required.

      “All we want to do is take existing pre-trained models and have them successfully interface with each other,” says Anurag Ajay, a PhD student in the MIT Department of Electrical Engineering and Computer Science (EECS) and a CSAIL affiliate. “Instead of pushing for one model to do everything, we combine multiple ones that leverage different modalities of internet data. When used in tandem, they help with robotic decision-making and can potentially aid with tasks in homes, factories, and construction sites.”

      These models also need some form of “eyes” to understand the environment they’re operating in and correctly execute each sub-goal. The team used a large video diffusion model to augment the initial planning completed by the LLM, which collects geometric and physical information about the world from footage on the internet. In turn, the video model generates an observation trajectory plan, refining the LLM’s outline to incorporate new physical knowledge.This process, known as iterative refinement, allows HiP to reason about its ideas, taking in feedback at each stage to generate a more practical outline. The flow of feedback is similar to writing an article, where an author may send their draft to an editor, and with those revisions incorporated in, the publisher reviews for any last changes and finalizes.

      In this case, the top of the hierarchy is an egocentric action model, or a sequence of first-person images that infer which actions should take place based on its surroundings. During this stage, the observation plan from the video model is mapped over the space visible to the robot, helping the machine decide how to execute each task within the long-horizon goal. If a robot uses HiP to make tea, this means it will have mapped out exactly where the pot, sink, and other key visual elements are, and begin completing each sub-goal.Still, the multimodal work is limited by the lack of high-quality video foundation models. Once available, they could interface with HiP’s small-scale video models to further enhance visual sequence prediction and robot action generation. A higher-quality version would also reduce the current data requirements of the video models.That being said, the CSAIL team’s approach only used a tiny bit of data overall. Moreover, HiP was cheap to train and demonstrated the potential of using readily available foundation models to complete long-horizon tasks. “What Anurag has demonstrated is proof-of-concept of how we can take models trained on separate tasks and data modalities and combine them into models for robotic planning. In the future, HiP could be augmented with pre-trained models that can process touch and sound to make better plans,” says senior author Pulkit Agrawal, MIT assistant professor in EECS and director of the Improbable AI Lab. The group is also considering applying HiP to solving real-world long-horizon tasks in robotics.Ajay and Agrawal are lead authors on a paper describing the work. They are joined by MIT professors and CSAIL principal investigators Tommi Jaakkola, Joshua Tenenbaum, and Leslie Pack Kaelbling; CSAIL research affiliate and MIT-IBM AI Lab research manager Akash Srivastava; graduate students Seungwook Han and Yilun Du ’19; former postdoc Abhishek Gupta, who is now assistant professor at University of Washington; and former graduate student Shuang Li PhD ’23.

      The team’s work was supported, in part, by the National Science Foundation, the U.S. Defense Advanced Research Projects Agency, the U.S. Army Research Office, the U.S. Office of Naval Research Multidisciplinary University Research Initiatives, and the MIT-IBM Watson AI Lab. Their findings were presented at the 2023 Conference on Neural Information Processing Systems (NeurIPS). More

    • 113 Shares119 Views

      in Data Management & Statistics

      Co-creating climate futures with real-time data and spatial storytelling

      by

      Virtual story worlds and game engines aren’t just for video games anymore. They are now tools for scientists and storytellers to digitally twin existing physical spaces and then turn them into vessels to dream up speculative climate stories and build collective designs of the future. That’s the theory and practice behind the MIT WORLDING initiative.

      Twice this year, WORLDING matched world-class climate story teams working in XR (extended reality) with relevant labs and researchers across MIT. One global group returned for a virtual gathering online in partnership with Unity for Humanity, while another met for one weekend in person, hosted at the MIT Media Lab.

      “We are witnessing the birth of an emergent field that fuses climate science, urban planning, real-time 3D engines, nonfiction storytelling, and speculative fiction, and it is all fueled by the urgency of the climate crises,” says Katerina Cizek, lead designer of the WORLDING initiative at the Co-Creation Studio of MIT Open Documentary Lab. “Interdisciplinary teams are forming and blossoming around the planet to collectively imagine and tell stories of healthy, livable worlds in virtual 3D spaces and then finding direct ways to translate that back to earth, literally.”

      At this year’s virtual version of WORLDING, five multidisciplinary teams were selected from an open call. In a week-long series of research and development gatherings, the teams met with MIT scientists, staff, fellows, students, and graduates, as well as other leading figures in the field. Guests ranged from curators at film festivals such as Sundance and Venice, climate policy specialists, and award-winning media creators to software engineers and renowned Earth and atmosphere scientists. The teams heard from MIT scholars in diverse domains, including geomorphology, urban planning as acts of democracy, and climate researchers at MIT Media Lab.

      Mapping climate data

      “We are measuring the Earth’s environment in increasingly data-driven ways. Hundreds of terabytes of data are taken every day about our planet in order to study the Earth as a holistic system, so we can address key questions about global climate change,” explains Rachel Connolly, an MIT Media Lab research scientist focused in the “Future Worlds” research theme, in a talk to the group. “Why is this important for your work and storytelling in general? Having the capacity to understand and leverage this data is critical for those who wish to design for and successfully operate in the dynamic Earth environment.”

      Making sense of billions of data points was a key theme during this year’s sessions. In another talk, Taylor Perron, an MIT professor of Earth, atmospheric and planetary sciences, shared how his team uses computational modeling combined with many other scientific processes to better understand how geology, climate, and life intertwine to shape the surfaces of Earth and other planets. His work resonated with one WORLDING team in particular, one aiming to digitally reconstruct the pre-Hispanic Lake Texcoco — where current day Mexico City is now situated — as a way to contrast and examine the region’s current water crisis.

      Democratizing the future

      While WORLDING approaches rely on rigorous science and the interrogation of large datasets, they are also founded on democratizing community-led approaches.

      MIT Department of Urban Studies and Planning graduate Lafayette Cruise MCP ’19 met with the teams to discuss how he moved his own practice as a trained urban planner to include a futurist component involving participatory methods. “I felt we were asking the same limited questions in regards to the future we were wanting to produce. We’re very limited, very constrained, as to whose values and comforts are being centered. There are so many possibilities for how the future could be.”

      Scaling to reach billions

      This work scales from the very local to massive global populations. Climate policymakers are concerned with reaching billions of people in the line of fire. “We have a goal to reach 1 billion people with climate resilience solutions,” says Nidhi Upadhyaya, deputy director at Atlantic Council’s Adrienne Arsht-Rockefeller Foundation Resilience Center. To get that reach, Upadhyaya is turning to games. “There are 3.3 billion-plus people playing video games across the world. Half of these players are women. This industry is worth $300 billion. Africa is currently among the fastest-growing gaming markets in the world, and 55 percent of the global players are in the Asia Pacific region.” She reminded the group that this conversation is about policy and how formats of mass communication can be used for policymaking, bringing about change, changing behavior, and creating empathy within audiences.

      Socially engaged game development is also connected to education at Unity Technologies, a game engine company. “We brought together our education and social impact work because we really see it as a critical flywheel for our business,” said Jessica Lindl, vice president and global head of social impact/education at Unity Technologies, in the opening talk of WORLDING. “We upscale about 900,000 students, in university and high school programs around the world, and about 800,000 adults who are actively learning and reskilling and upskilling in Unity. Ultimately resulting in our mission of the ‘world is a better place with more creators in it,’ millions of creators who reach billions of consumers — telling the world stories, and fostering a more inclusive, sustainable, and equitable world.”

      Access to these technologies is key, especially the hardware. “Accessibility has been missing in XR,” explains Reginé Gilbert, who studies and teaches accessibility and disability in user experience design at New York University. “XR is being used in artificial intelligence, assistive technology, business, retail, communications, education, empathy, entertainment, recreation, events, gaming, health, rehabilitation meetings, navigation, therapy, training, video programming, virtual assistance wayfinding, and so many other uses. This is a fun fact for folks: 97.8 percent of the world hasn’t tried VR [virtual reality] yet, actually.”

      Meanwhile, new hardware is on its way. The WORLDING group got early insights into the highly anticipated Apple Vision Pro headset, which promises to integrate many forms of XR and personal computing in one device. “They’re really pushing this kind of pass-through or mixed reality,” said Dan Miller, a Unity engineer on the poly spatial team, collaborating with Apple, who described the experience of the device as “You are viewing the real world. You’re pulling up windows, you’re interacting with content. It’s a kind of spatial computing device where you have multiple apps open, whether it’s your email client next to your messaging client with a 3D game in the middle. You’re interacting with all these things in the same space and at different times.”

      “WORLDING combines our passion for social-impact storytelling and incredible innovative storytelling,” said Paisley Smith of the Unity for Humanity Program at Unity Technologies. She added, “This is an opportunity for creators to incubate their game-changing projects and connect with experts across climate, story, and technology.”

      Meeting at MIT

      In a new in-person iteration of WORLDING this year, organizers collaborated closely with Connolly at the MIT Media Lab to co-design an in-person weekend conference Oct. 25 – Nov. 7 with 45 scholars and professionals who visualize climate data at NASA, the National Oceanic and Atmospheric Administration, planetariums, and museums across the United States.

      A participant said of the event, “An incredible workshop that had had a profound effect on my understanding of climate data storytelling and how to combine different components together for a more [holistic] solution.”

      “With this gathering under our new Future Worlds banner,” says Dava Newman, director of the MIT Media Lab and Apollo Program Professor of Astronautics chair, “the Media Lab seeks to affect human behavior and help societies everywhere to improve life here on Earth and in worlds beyond, so that all — the sentient, natural, and cosmic — worlds may flourish.” 

      “WORLDING’s virtual-only component has been our biggest strength because it has enabled a true, international cohort to gather, build, and create together. But this year, an in-person version showed broader opportunities that spatial interactivity generates — informal Q&As, physical worksheets, and larger-scale ideation, all leading to deeper trust-building,” says WORLDING producer Srushti Kamat SM ’23.

      The future and potential of WORLDING lies in the ongoing dialogue between the virtual and physical, both in the work itself and in the format of the workshops. More

    • 125 Shares169 Views

      in Data Management & Statistics

      Technique could efficiently solve partial differential equations for numerous applications

      by

      In fields such as physics and engineering, partial differential equations (PDEs) are used to model complex physical processes to generate insight into how some of the most complicated physical and natural systems in the world function.

      To solve these difficult equations, researchers use high-fidelity numerical solvers, which can be very time-consuming and computationally expensive to run. The current simplified alternative, data-driven surrogate models, compute the goal property of a solution to PDEs rather than the whole solution. Those are trained on a set of data that has been generated by the high-fidelity solver, to predict the output of the PDEs for new inputs. This is data-intensive and expensive because complex physical systems require a large number of simulations to generate enough data. 

      In a new paper, “Physics-enhanced deep surrogates for partial differential equations,” published in December in Nature Machine Intelligence, a new method is proposed for developing data-driven surrogate models for complex physical systems in such fields as mechanics, optics, thermal transport, fluid dynamics, physical chemistry, and climate models.

      The paper was authored by MIT’s professor of applied mathematics Steven G. Johnson along with Payel Das and Youssef Mroueh of the MIT-IBM Watson AI Lab and IBM Research; Chris Rackauckas of Julia Lab; and Raphaël Pestourie, a former MIT postdoc who is now at Georgia Tech. The authors call their method “physics-enhanced deep surrogate” (PEDS), which combines a low-fidelity, explainable physics simulator with a neural network generator. The neural network generator is trained end-to-end to match the output of the high-fidelity numerical solver.

      “My aspiration is to replace the inefficient process of trial and error with systematic, computer-aided simulation and optimization,” says Pestourie. “Recent breakthroughs in AI like the large language model of ChatGPT rely on hundreds of billions of parameters and require vast amounts of resources to train and evaluate. In contrast, PEDS is affordable to all because it is incredibly efficient in computing resources and has a very low barrier in terms of infrastructure needed to use it.”

      In the article, they show that PEDS surrogates can be up to three times more accurate than an ensemble of feedforward neural networks with limited data (approximately 1,000 training points), and reduce the training data needed by at least a factor of 100 to achieve a target error of 5 percent. Developed using the MIT-designed Julia programming language, this scientific machine-learning method is thus efficient in both computing and data.

      The authors also report that PEDS provides a general, data-driven strategy to bridge the gap between a vast array of simplified physical models with corresponding brute-force numerical solvers modeling complex systems. This technique offers accuracy, speed, data efficiency, and physical insights into the process.

      Says Pestourie, “Since the 2000s, as computing capabilities improved, the trend of scientific models has been to increase the number of parameters to fit the data better, sometimes at the cost of a lower predictive accuracy. PEDS does the opposite by choosing its parameters smartly. It leverages the technology of automatic differentiation to train a neural network that makes a model with few parameters accurate.”

      “The main challenge that prevents surrogate models from being used more widely in engineering is the curse of dimensionality — the fact that the needed data to train a model increases exponentially with the number of model variables,” says Pestourie. “PEDS reduces this curse by incorporating information from the data and from the field knowledge in the form of a low-fidelity model solver.”

      The researchers say that PEDS has the potential to revive a whole body of the pre-2000 literature dedicated to minimal models — intuitive models that PEDS could make more accurate while also being predictive for surrogate model applications.

      “The application of the PEDS framework is beyond what we showed in this study,” says Das. “Complex physical systems governed by PDEs are ubiquitous, from climate modeling to seismic modeling and beyond. Our physics-inspired fast and explainable surrogate models will be of great use in those applications, and play a complementary role to other emerging techniques, like foundation models.”

      The research was supported by the MIT-IBM Watson AI Lab and the U.S. Army Research Office through the Institute for Soldier Nanotechnologies.  More

    • 63 Shares189 Views

      in Data Management & Statistics

      “MIT can give you ‘superpowers’”

      by

      Speaking at the virtual MITx MicroMasters Program Joint Completion Celebration last summer, Diogo da Silva Branco Magalhães described watching a Spider-Man movie with his 8-year-old son and realizing that his son thought MIT was a fictional entity that existed only in the Marvel universe.

      “I had to tell him that MIT also exists in the real world, and that some of the programs are available online for everyone,” says da Silva Branco Magalhães, who earned his credential in the MicroMasters in Statistics and Data Science program. “You don’t need to be a superhero to participate in an MIT program, but MIT can give you ‘superpowers.’ In my case, the superpower that I was looking to acquire was a better understanding of the key technologies that are shaping the future of transportation.

      Part of MIT Open Learning, the MicroMasters programs have drawn in almost 1.4 million learners, spanning nearly every country in the world. More than 7,500 people have earned their credentials across the MicroMasters programs, including: Statistics and Data Science; Supply Chain Management; Data, Economics, and Design of Policy; Principles of Manufacturing; and Finance. 

      Earning his MicroMasters credential not only gave da Silva Branco Magalhães a strong foundation to tackle more complex transportation problems, but it also opened the door to pursuing an accelerated graduate degree via a Northwestern University online program.

      Learners who earn their MicroMasters credentials gain the opportunity to apply to and continue their studies at a pathway school. The MicroMasters in Statistics and Data Science credential can be applied as credit for a master’s program at more than 30 universities, as well as MIT’s PhD Program in Social and Engineering Systems. Da Silva Branco Magalhães, originally from Portugal and now based in Australia, seized this opportunity and enrolled in Northwestern University’s Master’s in Data Science for MIT MicroMasters Credential Holders. 

      The pathway to an enhanced career

      The pathway model launched in 2016 with the MicroMasters in Supply Chain Management. Now, there are over 50 pathway institutions that offer more than 100 different programs for master’s degrees. With pathway institutions located around the world, MicroMasters credential holders can obtain master’s degrees from local residential or virtual programs, at a location convenient to them. They can receive credit for their MicroMasters courses upon acceptance, providing flexibility for online programs and also shortening the time needed on site for residential programs.

      “The pathways expand opportunities for learners, and also help universities attract a broader range of potential students, which can enrich their programs,” says Dana Doyle, senior director for the MicroMasters Program at MIT Open Learning. “This is a tangible way we can achieve our mission of expanding education access.”

      Da Silva Branco Magalhães began the MicroMasters in Statistics and Data Science program in 2020, ultimately completing the program in 2022.

      “After having worked for 20 years in the transportation sector in various roles, I realized I was no longer equipped as a professional to deal with the new technologies that were set to disrupt the mobility sector,” says da Silva Branco Magalhães. “It became clear to me that data and AI were the driving forces behind new products and services such as autonomous vehicles, on-demand transport, or mobility as a service, but I didn’t really understand how data was being used to achieve these outcomes, so I needed to improve my knowledge.”

      July 2023 MicroMasters Program Joint Completion Celebration for SCM, DEDP, PoM, SDS, and FinVideo: MIT Open Learning

      The MicroMasters in Statistics and Data Science was developed by the MIT Institute for Data, Systems, and Society and MITx. Credential holders are required to complete four courses equivalent to graduate-level courses in statistics and data science at MIT and a capstone exam comprising four two-hour proctored exams.

      “The content is world-class,” da Silva Branco Magalhães says of the program. “Even the most complex concepts were explained in a very intuitive way. The exercises and the capstone exam are challenging and stimulating — and MIT-level — which makes this credential highly valuable in the market.”

      Da Silva Branco Magalhães also found the discussion forum very useful, and valued conversations with his colleagues, noting that many of these discussions later continued after completion of the program.

      Gaining analysis and leadership skills

      Now in the Northwestern pathway program, da Silva Branco Magalhães finds that the MicroMasters in Statistics and Data Science program prepared him well for this next step in his studies. The nine-course, accelerated, online master’s program is designed to offer the same depth and rigor of Northwestern’s 12-course MS in Data Science program, aiming to help students build essential analysis and leadership skills that can be directly implemented into the professional realm. Students learn how to make reliable predictions using traditional statistics and machine learning methods.

      Da Silva Branco Magalhães says he has appreciated the remote nature of the Northwestern program, as he started it in France and then completed the first three courses in Australia. He also values the high number of elective courses, allowing students to design the master’s program according to personal preferences and interests.

      “I want to be prepared to meet the challenges and seize the opportunities that AI and data science technologies will bring to the professional realm,” he says. “With this credential, there are no limits to what you can achieve in the field of data science.” More

    • 113 Shares119 Views

      in Data Management & Statistics

      Image recognition accuracy: An unseen challenge confounding today’s AI

      by

      Imagine you are scrolling through the photos on your phone and you come across an image that at first you can’t recognize. It looks like maybe something fuzzy on the couch; could it be a pillow or a coat? After a couple of seconds it clicks — of course! That ball of fluff is your friend’s cat, Mocha. While some of your photos could be understood in an instant, why was this cat photo much more difficult?

      MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) researchers were surprised to find that despite the critical importance of understanding visual data in pivotal areas ranging from health care to transportation to household devices, the notion of an image’s recognition difficulty for humans has been almost entirely ignored. One of the major drivers of progress in deep learning-based AI has been datasets, yet we know little about how data drives progress in large-scale deep learning beyond that bigger is better.

      In real-world applications that require understanding visual data, humans outperform object recognition models despite the fact that models perform well on current datasets, including those explicitly designed to challenge machines with debiased images or distribution shifts. This problem persists, in part, because we have no guidance on the absolute difficulty of an image or dataset. Without controlling for the difficulty of images used for evaluation, it’s hard to objectively assess progress toward human-level performance, to cover the range of human abilities, and to increase the challenge posed by a dataset.

      To fill in this knowledge gap, David Mayo, an MIT PhD student in electrical engineering and computer science and a CSAIL affiliate, delved into the deep world of image datasets, exploring why certain images are more difficult for humans and machines to recognize than others. “Some images inherently take longer to recognize, and it’s essential to understand the brain’s activity during this process and its relation to machine learning models. Perhaps there are complex neural circuits or unique mechanisms missing in our current models, visible only when tested with challenging visual stimuli. This exploration is crucial for comprehending and enhancing machine vision models,” says Mayo, a lead author of a new paper on the work.

      This led to the development of a new metric, the “minimum viewing time” (MVT), which quantifies the difficulty of recognizing an image based on how long a person needs to view it before making a correct identification. Using a subset of ImageNet, a popular dataset in machine learning, and ObjectNet, a dataset designed to test object recognition robustness, the team showed images to participants for varying durations from as short as 17 milliseconds to as long as 10 seconds, and asked them to choose the correct object from a set of 50 options. After over 200,000 image presentation trials, the team found that existing test sets, including ObjectNet, appeared skewed toward easier, shorter MVT images, with the vast majority of benchmark performance derived from images that are easy for humans.

      The project identified interesting trends in model performance — particularly in relation to scaling. Larger models showed considerable improvement on simpler images but made less progress on more challenging images. The CLIP models, which incorporate both language and vision, stood out as they moved in the direction of more human-like recognition.

      “Traditionally, object recognition datasets have been skewed towards less-complex images, a practice that has led to an inflation in model performance metrics, not truly reflective of a model’s robustness or its ability to tackle complex visual tasks. Our research reveals that harder images pose a more acute challenge, causing a distribution shift that is often not accounted for in standard evaluations,” says Mayo. “We released image sets tagged by difficulty along with tools to automatically compute MVT, enabling MVT to be added to existing benchmarks and extended to various applications. These include measuring test set difficulty before deploying real-world systems, discovering neural correlates of image difficulty, and advancing object recognition techniques to close the gap between benchmark and real-world performance.”

      “One of my biggest takeaways is that we now have another dimension to evaluate models on. We want models that are able to recognize any image even if — perhaps especially if — it’s hard for a human to recognize. We’re the first to quantify what this would mean. Our results show that not only is this not the case with today’s state of the art, but also that our current evaluation methods don’t have the ability to tell us when it is the case because standard datasets are so skewed toward easy images,” says Jesse Cummings, an MIT graduate student in electrical engineering and computer science and co-first author with Mayo on the paper.

      From ObjectNet to MVT

      A few years ago, the team behind this project identified a significant challenge in the field of machine learning: Models were struggling with out-of-distribution images, or images that were not well-represented in the training data. Enter ObjectNet, a dataset comprised of images collected from real-life settings. The dataset helped illuminate the performance gap between machine learning models and human recognition abilities, by eliminating spurious correlations present in other benchmarks — for example, between an object and its background. ObjectNet illuminated the gap between the performance of machine vision models on datasets and in real-world applications, encouraging use for many researchers and developers — which subsequently improved model performance.

      Fast forward to the present, and the team has taken their research a step further with MVT. Unlike traditional methods that focus on absolute performance, this new approach assesses how models perform by contrasting their responses to the easiest and hardest images. The study further explored how image difficulty could be explained and tested for similarity to human visual processing. Using metrics like c-score, prediction depth, and adversarial robustness, the team found that harder images are processed differently by networks. “While there are observable trends, such as easier images being more prototypical, a comprehensive semantic explanation of image difficulty continues to elude the scientific community,” says Mayo.

      In the realm of health care, for example, the pertinence of understanding visual complexity becomes even more pronounced. The ability of AI models to interpret medical images, such as X-rays, is subject to the diversity and difficulty distribution of the images. The researchers advocate for a meticulous analysis of difficulty distribution tailored for professionals, ensuring AI systems are evaluated based on expert standards, rather than layperson interpretations.

      Mayo and Cummings are currently looking at neurological underpinnings of visual recognition as well, probing into whether the brain exhibits differential activity when processing easy versus challenging images. The study aims to unravel whether complex images recruit additional brain areas not typically associated with visual processing, hopefully helping demystify how our brains accurately and efficiently decode the visual world.

      Toward human-level performance

      Looking ahead, the researchers are not only focused on exploring ways to enhance AI’s predictive capabilities regarding image difficulty. The team is working on identifying correlations with viewing-time difficulty in order to generate harder or easier versions of images.

      Despite the study’s significant strides, the researchers acknowledge limitations, particularly in terms of the separation of object recognition from visual search tasks. The current methodology does concentrate on recognizing objects, leaving out the complexities introduced by cluttered images.

      “This comprehensive approach addresses the long-standing challenge of objectively assessing progress towards human-level performance in object recognition and opens new avenues for understanding and advancing the field,” says Mayo. “With the potential to adapt the Minimum Viewing Time difficulty metric for a variety of visual tasks, this work paves the way for more robust, human-like performance in object recognition, ensuring that models are truly put to the test and are ready for the complexities of real-world visual understanding.”

      “This is a fascinating study of how human perception can be used to identify weaknesses in the ways AI vision models are typically benchmarked, which overestimate AI performance by concentrating on easy images,” says Alan L. Yuille, Bloomberg Distinguished Professor of Cognitive Science and Computer Science at Johns Hopkins University, who was not involved in the paper. “This will help develop more realistic benchmarks leading not only to improvements to AI but also make fairer comparisons between AI and human perception.” 

      “It’s widely claimed that computer vision systems now outperform humans, and on some benchmark datasets, that’s true,” says Anthropic technical staff member Simon Kornblith PhD ’17, who was also not involved in this work. “However, a lot of the difficulty in those benchmarks comes from the obscurity of what’s in the images; the average person just doesn’t know enough to classify different breeds of dogs. This work instead focuses on images that people can only get right if given enough time. These images are generally much harder for computer vision systems, but the best systems are only a bit worse than humans.”

      Mayo, Cummings, and Xinyu Lin MEng ’22 wrote the paper alongside CSAIL Research Scientist Andrei Barbu, CSAIL Principal Research Scientist Boris Katz, and MIT-IBM Watson AI Lab Principal Researcher Dan Gutfreund. The researchers are affiliates of the MIT Center for Brains, Minds, and Machines.

      The team is presenting their work at the 2023 Conference on Neural Information Processing Systems (NeurIPS). More

    • 125 Shares129 Views

      in Data Management & Statistics

      Three MIT students selected as inaugural MIT-Pillar AI Collective Fellows

      by

      MIT-Pillar AI Collective has announced three inaugural fellows for the fall 2023 semester. With support from the program, the graduate students, who are in their final year of a master’s or PhD program, will conduct research in the areas of artificial intelligence, machine learning, and data science with the aim of commercializing their innovations.

      Launched by MIT’s School of Engineering and Pillar VC in 2022, the MIT-Pillar AI Collective supports faculty, postdocs, and students conducting research on AI, machine learning, and data science. Supported by a gift from Pillar VC and administered by the MIT Deshpande Center for Technological Innovation, the mission of the program is to advance research toward commercialization.

      The fall 2023 MIT-Pillar AI Collective Fellows are:

      Alexander Andonian SM ’21 is a PhD candidate in electrical engineering and computer science whose research interests lie in computer vision, deep learning, and artificial intelligence. More specifically, he is focused on building a generalist, multimodal AI scientist driven by generative vision-language model agents capable of proposing scientific hypotheses, running computational experiments, evaluating supporting evidence, and verifying conclusions in the same way as a human researcher or reviewer. Such an agent could be trained to optimally distill and communicate its findings for human consumption and comprehension. Andonian’s work holds the promise of creating a concrete foundation for rigorously building and holistically testing the next-generation autonomous AI agent for science. In addition to his research, Andonian is the CEO and co-founder of Reelize, a startup that offers a generative AI video tool that effortlessly turns long videos into short clips — and originated from his business coursework and was supported by MIT Sandbox. Andonian is also a founding AI researcher at Poly AI, an early-stage YC-backed startup building AI design tools. Andonian earned an SM from MIT and a BS in neuroscience, physics, and mathematics from Bates College.

      Daniel Magley is a PhD candidate in the Harvard-MIT Program in Health Sciences and Technology who is passionate about making a healthy, fully functioning mind and body a reality for all. His leading-edge research is focused on developing a swallowable wireless thermal imaging capsule that could be used in treating and monitoring inflammatory bowel diseases and their manifestations, such as Crohn’s disease. Providing increased sensitivity and eliminating the need for bowel preparation, the capsule has the potential to vastly improve treatment efficacy and overall patient experience in routine monitoring. The capsule has completed animal studies and is entering human studies at Mass General Brigham, where Magley leads a team of engineers in the hospital’s largest translational research lab, the Tearney Lab. Following the human pilot studies, the largest technological and regulatory risks will be cleared for translation. Magley will then begin focusing on a multi-site study to get the device into clinics, with the promise of benefiting patients across the country. Magley earned a BS in electrical engineering from Caltech.

      Madhumitha Ravichandra is a PhD candidate interested in advancing heat transfer and surface engineering techniques to enhance the safety and performance of nuclear energy systems and reduce their environmental impacts. Leveraging her deep knowledge of the integration of explainable AI with high-throughput autonomous experimentation, she seeks to transform the development of radiation-hardened (rad-hard) sensors, which could potentially withstand and function amidst radiation levels that would render conventional sensors useless. By integrating explainable AI with high-throughput autonomous experimentation, she aims to rapidly iterate designs, test under varied conditions, and ensure that the final product is both robust and transparent in its operations. Her work in this space could shift the paradigm in rad-hard sensor development, addressing a glaring void in the market and redefining standards, ensuring that nuclear and space applications are safer, more efficient, and at the cutting edge of technological progress. Ravichandran earned a BTech in mechanical engineering from SASTRA University, India. More

    • 138 Shares169 Views

      in Data Management & Statistics

      Automated system teaches users when to collaborate with an AI assistant

      by

      Artificial intelligence models that pick out patterns in images can often do so better than human eyes — but not always. If a radiologist is using an AI model to help her determine whether a patient’s X-rays show signs of pneumonia, when should she trust the model’s advice and when should she ignore it?

      A customized onboarding process could help this radiologist answer that question, according to researchers at MIT and the MIT-IBM Watson AI Lab. They designed a system that teaches a user when to collaborate with an AI assistant.

      In this case, the training method might find situations where the radiologist trusts the model’s advice — except she shouldn’t because the model is wrong. The system automatically learns rules for how she should collaborate with the AI, and describes them with natural language.

      During onboarding, the radiologist practices collaborating with the AI using training exercises based on these rules, receiving feedback about her performance and the AI’s performance.

      The researchers found that this onboarding procedure led to about a 5 percent improvement in accuracy when humans and AI collaborated on an image prediction task. Their results also show that just telling the user when to trust the AI, without training, led to worse performance.

      Importantly, the researchers’ system is fully automated, so it learns to create the onboarding process based on data from the human and AI performing a specific task. It can also adapt to different tasks, so it can be scaled up and used in many situations where humans and AI models work together, such as in social media content moderation, writing, and programming.

      “So often, people are given these AI tools to use without any training to help them figure out when it is going to be helpful. That’s not what we do with nearly every other tool that people use — there is almost always some kind of tutorial that comes with it. But for AI, this seems to be missing. We are trying to tackle this problem from a methodological and behavioral perspective,” says Hussein Mozannar, a graduate student in the Social and Engineering Systems doctoral program within the Institute for Data, Systems, and Society (IDSS) and lead author of a paper about this training process.

      The researchers envision that such onboarding will be a crucial part of training for medical professionals.

      “One could imagine, for example, that doctors making treatment decisions with the help of AI will first have to do training similar to what we propose. We may need to rethink everything from continuing medical education to the way clinical trials are designed,” says senior author David Sontag, a professor of EECS, a member of the MIT-IBM Watson AI Lab and the MIT Jameel Clinic, and the leader of the Clinical Machine Learning Group of the Computer Science and Artificial Intelligence Laboratory (CSAIL).

      Mozannar, who is also a researcher with the Clinical Machine Learning Group, is joined on the paper by Jimin J. Lee, an undergraduate in electrical engineering and computer science; Dennis Wei, a senior research scientist at IBM Research; and Prasanna Sattigeri and Subhro Das, research staff members at the MIT-IBM Watson AI Lab. The paper will be presented at the Conference on Neural Information Processing Systems.

      Training that evolves

      Existing onboarding methods for human-AI collaboration are often composed of training materials produced by human experts for specific use cases, making them difficult to scale up. Some related techniques rely on explanations, where the AI tells the user its confidence in each decision, but research has shown that explanations are rarely helpful, Mozannar says.

      “The AI model’s capabilities are constantly evolving, so the use cases where the human could potentially benefit from it are growing over time. At the same time, the user’s perception of the model continues changing. So, we need a training procedure that also evolves over time,” he adds.

      To accomplish this, their onboarding method is automatically learned from data. It is built from a dataset that contains many instances of a task, such as detecting the presence of a traffic light from a blurry image.

      The system’s first step is to collect data on the human and AI performing this task. In this case, the human would try to predict, with the help of AI, whether blurry images contain traffic lights.

      The system embeds these data points onto a latent space, which is a representation of data in which similar data points are closer together. It uses an algorithm to discover regions of this space where the human collaborates incorrectly with the AI. These regions capture instances where the human trusted the AI’s prediction but the prediction was wrong, and vice versa.

      Perhaps the human mistakenly trusts the AI when images show a highway at night.

      After discovering the regions, a second algorithm utilizes a large language model to describe each region as a rule, using natural language. The algorithm iteratively fine-tunes that rule by finding contrasting examples. It might describe this region as “ignore AI when it is a highway during the night.”

      These rules are used to build training exercises. The onboarding system shows an example to the human, in this case a blurry highway scene at night, as well as the AI’s prediction, and asks the user if the image shows traffic lights. The user can answer yes, no, or use the AI’s prediction.

      If the human is wrong, they are shown the correct answer and performance statistics for the human and AI on these instances of the task. The system does this for each region, and at the end of the training process, repeats the exercises the human got wrong.

      “After that, the human has learned something about these regions that we hope they will take away in the future to make more accurate predictions,” Mozannar says.

      Onboarding boosts accuracy

      The researchers tested this system with users on two tasks — detecting traffic lights in blurry images and answering multiple choice questions from many domains (such as biology, philosophy, computer science, etc.).

      They first showed users a card with information about the AI model, how it was trained, and a breakdown of its performance on broad categories. Users were split into five groups: Some were only shown the card, some went through the researchers’ onboarding procedure, some went through a baseline onboarding procedure, some went through the researchers’ onboarding procedure and were given recommendations of when they should or should not trust the AI, and others were only given the recommendations.

      Only the researchers’ onboarding procedure without recommendations improved users’ accuracy significantly, boosting their performance on the traffic light prediction task by about 5 percent without slowing them down. However, onboarding was not as effective for the question-answering task. The researchers believe this is because the AI model, ChatGPT, provided explanations with each answer that convey whether it should be trusted.

      But providing recommendations without onboarding had the opposite effect — users not only performed worse, they took more time to make predictions.

      “When you only give someone recommendations, it seems like they get confused and don’t know what to do. It derails their process. People also don’t like being told what to do, so that is a factor as well,” Mozannar says.

      Providing recommendations alone could harm the user if those recommendations are wrong, he adds. With onboarding, on the other hand, the biggest limitation is the amount of available data. If there aren’t enough data, the onboarding stage won’t be as effective, he says.

      In the future, he and his collaborators want to conduct larger studies to evaluate the short- and long-term effects of onboarding. They also want to leverage unlabeled data for the onboarding process, and find methods to effectively reduce the number of regions without omitting important examples.

      “People are adopting AI systems willy-nilly, and indeed AI offers great potential, but these AI agents still sometimes makes mistakes. Thus, it’s crucial for AI developers to devise methods that help humans know when it’s safe to rely on the AI’s suggestions,” says Dan Weld, professor emeritus at the Paul G. Allen School of Computer Science and Engineering at the University of Washington, who was not involved with this research. “Mozannar et al. have created an innovative method for identifying situations where the AI is trustworthy, and (importantly) to describe them to people in a way that leads to better human-AI team interactions.”

      This work is funded, in part, by the MIT-IBM Watson AI Lab. More

    • 75 Shares149 Views

      in Data Management & Statistics

      AI accelerates problem-solving in complex scenarios

      by

      While Santa Claus may have a magical sleigh and nine plucky reindeer to help him deliver presents, for companies like FedEx, the optimization problem of efficiently routing holiday packages is so complicated that they often employ specialized software to find a solution.

      This software, called a mixed-integer linear programming (MILP) solver, splits a massive optimization problem into smaller pieces and uses generic algorithms to try and find the best solution. However, the solver could take hours — or even days — to arrive at a solution.

      The process is so onerous that a company often must stop the software partway through, accepting a solution that is not ideal but the best that could be generated in a set amount of time.

      Researchers from MIT and ETH Zurich used machine learning to speed things up.

      They identified a key intermediate step in MILP solvers that has so many potential solutions it takes an enormous amount of time to unravel, which slows the entire process. The researchers employed a filtering technique to simplify this step, then used machine learning to find the optimal solution for a specific type of problem.

      Their data-driven approach enables a company to use its own data to tailor a general-purpose MILP solver to the problem at hand.

      This new technique sped up MILP solvers between 30 and 70 percent, without any drop in accuracy. One could use this method to obtain an optimal solution more quickly or, for especially complex problems, a better solution in a tractable amount of time.

      This approach could be used wherever MILP solvers are employed, such as by ride-hailing services, electric grid operators, vaccination distributors, or any entity faced with a thorny resource-allocation problem.

      “Sometimes, in a field like optimization, it is very common for folks to think of solutions as either purely machine learning or purely classical. I am a firm believer that we want to get the best of both worlds, and this is a really strong instantiation of that hybrid approach,” says senior author Cathy Wu, the Gilbert W. Winslow Career Development Assistant Professor in Civil and Environmental Engineering (CEE), and a member of a member of the Laboratory for Information and Decision Systems (LIDS) and the Institute for Data, Systems, and Society (IDSS).

      Wu wrote the paper with co-lead authors Siriu Li, an IDSS graduate student, and Wenbin Ouyang, a CEE graduate student; as well as Max Paulus, a graduate student at ETH Zurich. The research will be presented at the Conference on Neural Information Processing Systems.

      Tough to solve

      MILP problems have an exponential number of potential solutions. For instance, say a traveling salesperson wants to find the shortest path to visit several cities and then return to their city of origin. If there are many cities which could be visited in any order, the number of potential solutions might be greater than the number of atoms in the universe.  

      “These problems are called NP-hard, which means it is very unlikely there is an efficient algorithm to solve them. When the problem is big enough, we can only hope to achieve some suboptimal performance,” Wu explains.

      An MILP solver employs an array of techniques and practical tricks that can achieve reasonable solutions in a tractable amount of time.

      A typical solver uses a divide-and-conquer approach, first splitting the space of potential solutions into smaller pieces with a technique called branching. Then, the solver employs a technique called cutting to tighten up these smaller pieces so they can be searched faster.

      Cutting uses a set of rules that tighten the search space without removing any feasible solutions. These rules are generated by a few dozen algorithms, known as separators, that have been created for different kinds of MILP problems. 

      Wu and her team found that the process of identifying the ideal combination of separator algorithms to use is, in itself, a problem with an exponential number of solutions.

      “Separator management is a core part of every solver, but this is an underappreciated aspect of the problem space. One of the contributions of this work is identifying the problem of separator management as a machine learning task to begin with,” she says.

      Shrinking the solution space

      She and her collaborators devised a filtering mechanism that reduces this separator search space from more than 130,000 potential combinations to around 20 options. This filtering mechanism draws on the principle of diminishing marginal returns, which says that the most benefit would come from a small set of algorithms, and adding additional algorithms won’t bring much extra improvement.

      Then they use a machine-learning model to pick the best combination of algorithms from among the 20 remaining options.

      This model is trained with a dataset specific to the user’s optimization problem, so it learns to choose algorithms that best suit the user’s particular task. Since a company like FedEx has solved routing problems many times before, using real data gleaned from past experience should lead to better solutions than starting from scratch each time.

      The model’s iterative learning process, known as contextual bandits, a form of reinforcement learning, involves picking a potential solution, getting feedback on how good it was, and then trying again to find a better solution.

      This data-driven approach accelerated MILP solvers between 30 and 70 percent without any drop in accuracy. Moreover, the speedup was similar when they applied it to a simpler, open-source solver and a more powerful, commercial solver.

      In the future, Wu and her collaborators want to apply this approach to even more complex MILP problems, where gathering labeled data to train the model could be especially challenging. Perhaps they can train the model on a smaller dataset and then tweak it to tackle a much larger optimization problem, she says. The researchers are also interested in interpreting the learned model to better understand the effectiveness of different separator algorithms.

      This research is supported, in part, by Mathworks, the National Science Foundation (NSF), the MIT Amazon Science Hub, and MIT’s Research Support Committee. More