Engineering tasks such as design, monitoring, inspection, and education and training rely on our ability to create high-fidelity models describing the behavior of structures in the physical world (e.g., bridges, wind turbines, pipelines, power transmission, etc.). Efficient execution of these engineering tasks in the information age relies on our ability to digitize the physical structures and systems and connect sensing data and physical models to the digital replica of the physical structure (digital twin). This study aims to create such digital representations for engineering systems, visualize them using virtual and augmented reality (VR/AR), and to study their suitability for performing engineering tasks.
To that end, 3D models of sample structures will be captured via laser scanners or developed in graphical software and connected to the physical structure using sensing data and existing mechanical models. These models will then be imported into mixed reality environment and enriched with data such as sensing, descriptions of condition, and other information of interest for the specific engineering task. Depending on the progress of the project, opportunities exist for participating in real tests of target structures instrumented with mechanical and optical sensors within the Newmark Structural Engineering Laboratory (NSEL) in the Department of Civil and Environmental Engineering. The end goal will be to create a prototype digital twin of a structure that can be viewed and interacted with using AR/VR headsets.
Student Background and Research Activities:
Successful applicants will have strong programming skills, and experience with game engines, virtual reality programming (e.g., Unity or similar platforms). No knowledge of civil, structural, and mechanical engineering is required. This project involves exciting research activities including mixed reality application development and programming, computer simulations, and potentially working with sensing systems. The student will also be working with mixed reality headsets such as the Meta Oculus, HTC Vive, and Magic Leap. Depending on the progress, this project may lead to a conference paper and/or a longer-term research position.
The Mechanics of Complex Systems Laboratory, headed by Professor Ahmed Elbanna, in the Grainger College of Engineering has an open REU position during Fall 2022.
Project Description:
Many natural hazards, such as earthquakes and landslides, occur suddenly with little or no warning, posing significant threat to lives and economy. These hazards usually involve physical processes that span several decades of spatial and temporal scales. An outstanding challenge is that these processes are largely hidden from direct observations and can be only inferred through indirect measurements. This limits our knowledge of the detailed mechanisms involved in these hazards, and thus, reduces our ability to control them or adequately mitigate their impact on the society and its infrastructure. Numerical modeling is emerging as the primary tool to bridge this observation data gap and to enable discovery of universal physics that transcends limited regional statistical rules. Our group has recently pioneered a hybrid finite element boundary element method (FEBE) that enables simulation of earthquake cycles in complex fault zones over long period of times, opening new opportunities for understanding the multiscale physics of crustal faults and seismicity. Further development of this method will enable building earthquake simulators on regional scales that will advance short- and long-term earthquake predictability and help improve seismic hazard models and resilience of the urban infrastructure.
Focus on REU:
The successful candidate will work with other members of the group on implementation of several programming assignments related to updating FEBE. Their specific task will include rewriting pieces of existing MATLAB/Python Finite Element Method codes into C++ and to develop an OpenMP/MPI implementation for parallelization of the C++ code. Duration and compensation: A total of $1500 for the semester. Expected workload is 10 hours/week for 10 weeks.
Skills required: The successful candidate should have some experience with numerical linear algebra and a working knowledge of C++ and MPI as applied in scientific computing.
The intern will work at the forefront of deep learning development for understanding the structure, properties, and physics of materials. Current deep learning approaches in materials science have generally ignored the angular symmetries in a material’s atomic structure, which is widely known in materials physics to be crucial for understanding a material’s properties. As such, towards enabling full understanding of materials by neural networks, the intern will study current models’ abilities to capture local and global atomic structures. The intern will then develop novel deep learning models to improve upon the previous state-of-the-art. Depending on the skills and interests of the intern, possible projects include:
Skills required:
Preferred skills:
In this project, we develop a data-model fusion framework that integrate IoT sensing, vehicle sensing, airborne sensing and satellite remote sensing with biophysical models and AI models to quantify soil organic carbon and nitrous oxide emission at field level and regional scale.
Modern scientific simulations have enabled us to study non-linear phenomena that are impossible to study otherwise. Among the most challenging problems is the study of Einstein's theory of relativity which predicts the existence of gravitational waves detected very recently be the LIGO collaboration. The Einstein Toolkit is a community driven framework for astrophysical simulations. I am interested in recruiting a student interested in imrpoving the automated testing framework in the Einstein Toolkit using GitHub.
This project will involve improving our continuous integration framework for the Einstein Toolkit using GitHub Actions and creating new HTML pages summarizing the result of each test.
The successful applicant will be involved with both the Gravity Group at NCSA and will be invited to participate in the weekly group meetings and discussions of their research projects.
Details: The Einstein Toolkit contains almost 300 regression tests to ensure that additions to the code do not introduce bugs that affect scientific results. The goal of this SPIN project is to build on results of a previous SPIN project and continue porting the testing system over to GitHub's Actions framework. This will involve familiarizing oneself with the GitHub Actions framework as well as understanding how to run the Einstein Toolkit test harness and interpret its results.
Preferred skills:
Contact Roland Haas
Modern scientific simulations have enabled us to study non-linear phenomena that are impossible to study otherwise. Among the most challenging problems is the study of Einstein's theory of relativity which predicts the existence of gravitational waves detected very recently be the LIGO collaboration. The Einstein Toolkit http://einsteintoolkit.org/ is a community driven framework for astrophysical simulations. I am interested in recruiting a student interested in improving the quality of gravitational waveform templates describing colliding black holes produced with the in the Einstein Toolkit.
This project will involve improving the ""parameter files"" used to set up binary black hole simulations of merging black holes used by the NCSA gravity group. This will involve running simulations on compute clusters at ACCESS and NCSA, visualizing simulation results using VisIt and matplotlib and analyze data using numpy.
The successful applicant will be involved with both the Gravity Group at NCSA and will be invited to participate in the weekly group meetings and discussions of their research projects.
Before applying, please do the Exercise described on https://wiki.ncsa.illinois.edu/display/~rhaas/SPIN+2022+-+2023+Exercise. I will not consider you for the project unless I have received the exercise!
Preferred Skills:
The selected student will use AI and high performance computing to study high dimensional signal manifolds that describe black holes that spin and precess and they spiral into each other and merge. The central aspect of this project is computational in nature and will involve the use of supercomputers to improve the convergence and accuracy of AI models as we use between a few thousand to tens of thousands of GPUs.
The student will use physics inspired neural operators to accelerate the solution of high-dimensional partial differential equations. The student will learn to use supercomputers to automate the design and training of physics inspired neural operators. 2) Automated AI workflows for big data experiments. The student will learn to create workflows that connect big data sources with remote AI repositories and HPC/edge/cloud computing resources. In particular, the student will learn how to publish models in the Data and Learning Hub for Science, and then connect supercomputers (Delta at NCSA, and Polaris and Aurora at Argonne) and data sources with distributed computing services such as funcX. These workflows will bypass human intervention to process massive datasets at scale. 3) Distributed AI inference and training. The student will learn how to train AI models at scale using between tens to thousands of GPUs, and then to port AI models into accelerated AI engines for distributed inference on HPC platforms. Science drivers for this project include gravitational wave astrophysics, cosmology and high energy physics.
There are a plethora of available methods to solve partial differential equations (PDEs). These approaches have been refined over many decades. However, PDEs that we encounter in real world problems typically demand the combination of computationally inefficient and time-consuming solvers. The selected student to work on this project will learn to use deep learning to accelerate the solution of PDEs in high dimensions for a variety of physical processes, with a particular focus in turbulence, weather modeling and finance.
Recent developments in artificial intelligence and big data have changed the landscape of technologies used by researchers to advance their scientific endeavors. These technologies include new software, hardware, cloud services, and programming frameworks that have rapidly grown in complexity. This changing landscape has caused a gap between the knowledge it takes to use these technologies and the knowledge it takes to shape them accordingly to specific use cases. Traditional training on using high-performance computing resources routinely offered to researchers does not address this gap, making it necessary to conduct a needs analysis to identify the training researchers need to use these technologies effectively.
This project aims to map and characterize this technology landscape and identify what training researchers need to use the technology effectively. The SPIN student will use various methods to collect information about the computing technologies and resources available, how researchers use them, and the knowledge gaps prohibiting their use. Example collection methods include surveys, participation in scientific events held on campus, and interviews. The key to this effort is direct interaction with researchers across campus to collect results that are not biased by technology providers.
The ideal candidate has strong communication skills, a passion for science, and a willingness to work across and on the interface between multiple fields. At the end of the project, we expect that the SPIN student will have developed a comprehensive understanding of several research computing technologies and the ability to conceptualize mechanisms to effectively collect relevant information that can support policymaking, including data analytics techniques. Finally, as a bonus, the SPIN student will have created a solid and extensive network of researchers across campus that will certainly broaden their experience and the range of possibilities for their future career.
Preferred Skills:
As part of research into use of the funcX (https://funcx.org, a portable function-as-a-service) system, the student will prototype a simple web based interface to performance data already collected by funcX in an ad-hoc fashion, allowing funcX users access to basic plots of task information such as duration and throughput. This is intended to help flesh out understanding within the funcX project of which monitoring information is useful to end users, how it might be presented, and what other information might be useful to collect. This work will be done as part of the globally distributed funcX team, and will provide experience in distributed collaborative open source software development practices.
Skills Required:
An estimated 84 million persons are forcibly displaced worldwide, and at least 70% of these are living in conditions of extreme poverty. More efficient targeting mechanisms are needed to better identify vulnerable families who are most in need of humanitarian assistance. Traditional targeting models rely on a proxy means testing (PMT) approach, where support programs target refugee families whose estimated consumption falls below a certain threshold. Despite the method’s practicality, it provides limited insights, its predictions are not very accurate, and it can impact the targeting effectiveness and fairness. Alternatively, multidimensional approaches to assessing poverty are now being applied to the refugee context. Yet, they require extensive information that is often unavailable or costly. This project applies machine learning and geospatial methods to novel data collected from Syrian refugees in Lebanon to develop more effective and operationalizable targeting strategies that provide a reliable complementarity to current PMT and multidimensional methods. The insights from this project have important implications for humanitarian organizations seeking to improve current targeting mechanisms, especially given increasing poverty and displacement and limited humanitarian funding.
It is well established that minority populations are more prone to environmental in justice and associated health disparities. This project will analyze a data set that will potentially unravel these disparities. The students are required to have skills associated with advance statistical methods, R programming. Students with prior experience in machine learning techniques will be given priority.
Health disparities, be it racial, economic, rural-urban, gender- or age-based, have come to the forefront across the world. To elucidate the biological, social, economic and psychological mechanisms of health disparities, and to develop interventions that engage community in targeting these mechanisms to reduce health disparities, it is necessary to work with complex multidimensional datasets containing molecular, genetic and biometric information from individuals, plus their socioeconomic status, local environment/safety, degree of segregation, access to medical care/education, and levels of pollution. We are developing novel statistical and ML approaches to harmonize these heterogeneous data and detect important contributors to health disparities. We are aiming to develop predictive tools to identify populations at-risk for poor health outcomes, in order to help community services, reach out and bring in those individuals for treatment earlier.
Student Contributions
The REU student will work with NCSA computational scientists and faculty collaborators in areas of women’s health and infectious diseases, as well as the representatives of the public health district, to gather, prepare and analyze health-related data, and develop novel statistical and ML approaches.
The amount of observations correlating pore shapes of geomaterials to their flow properties is very limited because of the complexity of performing direct measurements at a lab and field scale levels. At the same time, microscale characterization techniques, such as imaging and porosimetry, can be utilized in a robust way on small specimens making this approach advantageous for the projects where comprehensive characterization is required with very limited material availability. To study the influence of particular features, e.g. fractures or microscale pores on the flow - we want to embed this data into machine learning models to predict characteristics of the rock formations at larger scales. 3D data sets containing the results of X-ray CT scans of mm long shale specimens contain more than a thousand of high resolution images that need to be interpreted to determine the average and dominant pore sizes and presences of fractures. This information can be enhanced with the porosimetry data to provide an interpretation of the pore contribution in grayscale images. This approach could significantly improve our ability to understand and predict the hydraulic response of low permeable rock.
There is a need for rapid and autonomous learning of the hydraulic response of the system directly from lab data. HAL DGX A100 System at NCSA is a tool that can provide robust processing of combined imaging and porosimetry data for large datasets and calculation of the flow properties. If the neural network is trained on a few porous rock examples where larger scale data is available – it will significantly enhance the predictive power of the model. The intern will work on the development of a neural network that will provide flow properties of porous rock from combined microimaging and porosimetry measurements. Developed algorithms are expected to significantly advance geomaterials’ property prediction, providing robust tools for assessing their behavior.
Skills Required:
A successful candidate will help the STEM Illinois Nobel project to create a community health worker curriculum for middle and high school students. STEM Illinois is deeply rooted in the historic mission of land-grant institutions, which is to democratize higher education and to address the world's most pressing societal challenges. However, over 150 years after the Morrill Act was passed, social inequality reflects the harsh lived experiences of racially marginalized groups. These inequalities are especially visible in the field of computer science where, despite decades of pipeline programming, the number of underrepresented students remains alarmingly low. The goal of the STEM Illinois project is, to create a unique ecosystem that will nurture future computer scientists in industry and the academy. We believe that these students will follow Illinois tradition and address pressing societal challenges as innovators and Nobel Prize winners.
We seek to increase the number of marginalized students majoring in computer science and medicine. The student will employ thier skills as a pre-med student to help us consider the topics to cover and the areas for innovations as we train community health workers. Innovation in training is especially relevant during COVID-19. At the beginning of the pandemic, the governor of New York deployed the Army Corps of Engineers to create a temporary hospital to address anticipated high levels of morbidity and mortality. We argue that another critical step in addressing the pandemic is to create a corps of community health workers to document the impact of COVID-19 and to share their resiliency tools in real time. These young community health workers will create new knowledge and engage in competitions to improve health and wellness using innovative motives. The student will serve as a mentor to Nobel participants.
The intellectual contributions of this project is the documentation of the overwhelming cost of COVID-19 on the social, emotional, physical and financial health of U.S. citizens. The most important research contribution will be the volumes of data that we collect on marginalized groups' resiliency strategies during a pandemic.
This project researches frameworks and workflows for speech-to-text recognition in order to facilitate live auto captioning and creation of standard caption files for use in live events and video editing, utilizing and enhancing speech-to-text HPC/cloud services and seeks to advance the state of the art in speech-to-text recognition. A successful candidate would need to have completed CS125 (Intro to Computer Science) or have equivalent experience.
This project is a subset of a NASA Astrophysics Data Analysis Program (ADAP) project aimed at creating several science-ready data products to help astronomers’ search the literature in new ways. This goal is being accomplished by extending the NASA Astrophysics Data System (ADS), known as an invaluable literature resource, into a series of data resources. One part of this project involves the “reading” of figure captions using Optical Character Recognition (OCR) from scanned article pages. A large source of error in the OCR process comes from artifacts present on scanned pages -- scan-effects such as warping, lighting gradients and dust can generate many misspellings. This project is focused on better understanding these types of effects using image processing and analysis to better clean old images before OCR processing AND to potentially generate artificial training data using "aged" images of newer, digitized documents.
NCSA's Advanced Visualization Lab (AVL) in collaboration with iSchool are looking for an undergraduate research intern to help with a research project that builds on the research of doctoral candidate Rezvaneh (Shadi) Rezapour and Professor Jana Diesner, which uses data mining and natural language processing techniques to study the effects of issue-focused documentary films on various audiences by analyzing reviews and comments on streaming media sites. This new research will focus specifically on science-themed documentaries that use computational science research in their science explanations. Student researchers would be responsible for working with mentors in iSchool (Professor Jill Naiman) and AVL to collect data from streaming sites and analyze the data using existing purpose-built software and developing new tools.
No skills required. Students will be trained to conduct the classification of text documentary reviews. Preferred: background in interdisciplinary research.
This project is a subset of a NASA Astrophysics Data Analysis Program (ADAP) project aimed at creating several science-ready data products to help astronomers' search the literature in new ways. This goal is being accomplished by extending the NASA Astrophysics Data System (ADS), known as an invaluable literature resource, into a series of data resources. One part of this process will be classifying the figures that appear in journal articles by their "type" (for astronomical literature, classes will include things like "images of the sky," "graphs," "simulations," etc). For this summer research project, a student will help with this image classification both by by hand and testing with machine learning methods in collaboration with Dr. Jill Naiman and/or a grad student (School of Information Sciences and NCSA). The main parts of the project will involve developing the codebook of image classifications so that citizen scientists can complete more classifications on a large scale and running the by hand classification scripts. Options to extend this by working on the UI for the classification scripts (in Python, and/or for the Zooniverse citizen science platform) and working with the machine learning methods for image classification are available for interested students.
Skills required:
Preferred skills:
This project is a subset of a NASA Astrophysics Data Analysis Program (ADAP) project aimed at creating several science-ready data products to help astronomers’ search the literature in new ways. This goal is being accomplished by extending the NASA Astrophysics Data System (ADS), known as an invaluable literature resource, into a series of data resources. One part of this project involves the “reading” of figure captions using Optical Character Recognition (OCR) from scanned article pages. These first “readings” using OCR-automated methods usually result in many misspellings that need to be corrected. For this research project, a student will work on building a “spell check” for this scanned data by training a machine learning model on the misspelled and correctly-spelled collection of words and phrases gathered by the OCR and non-OCR processes.
Preferred skills:
This project is an immediate application of high-performance computing (HPC) in response to the COVID-19 pandemic. In this project, a successful candidate will help with improving a model to adapt to a more complex real-world challenge. The current model has successfully simulated the situation of lock-down, mask enforcement, social distancing, and vaccine roll-out, with complex social-interaction and spatial structure. For the scientific aspect, we will further investigate the situation of co-existing variants and immune escape. For the computation aspect, a SPIN intern will get hands-on experience applying advanced algorithms on the ACCESS supercomputers.
Skills required:
Preferred skills:
Many disease states, particularly in psychiatry, neurology, and cardiology, are often overlooked in our healthcare systems due to treatment barriers and untimely diagnoses. New disease screening methods are necessary to address these problems. This project proposes developing automated disease screening techniques that can infer clinical states, such as anxiety and manic depressive disorders, using machine learning, modeling, and human speech and language data. The team will integrate models with Clowder to demonstrate the automated annotation of speech/language/health data.
The cell membrane environment is complex and challenging to model. The Pogorelov Lab at Illinois develops workflows that combining computational and experimental molecular data. We work in close collaboration with experimental labs. Addressed questions include investigations of fundamental mechanisms of membrane activity, structural dynamics of peripheral and transmembrane proteins, and development of drugs. Modeling approaches include classical molecular dynamics, quantum electronic structure, and quantum nuclear dynamics. These projects include development of workflows for modeling and analysis of the lipid interactions with proteins and ions that are vital for life of the cell. The qualified student should have experience with R/Python programming, use of Linux environment, and of NAMD molecular modeling software.
This project will implement new functionality to the web front-end for the chemical box model MusicBox. MusicBox is used by scientists to develop their understanding of chemical interactions in the atmosphere. MusicBox is also used in the classroom to demonstrate the effects of atmospheric chemical processes and their impacts on human health and climate change. The goal of this project is to extend the capabilities of MusicBox to develop a front-end for aerosol process modeling. This will include developing new UI components in the web interface, implementing back-end API endpoints, and writing new data visualization web components. The students will be part of a collaboration between the University of Illinois at the National Center for Atmospheric Research in Boulder, CO.
Skills required:
Preferred skills:
In order to illustrate the electron density experienced by a fast-moving ion through a material, we have previously produced static, time-dependent, and virtual reality based visualizations. In this project, we aim to explore the concept of sonification, to supplement these results with audio, providing the audience with better comprehension of what is shown in the pictures. Required skills include Matlab, Python, and/or C++ coding, previous experience with visualization, and with data analysis. This exciting project combines expertise in Materials Science/Condensed Matter Physics with Computer Science.
Project goal: use deep learning to develop a better model capable of predicting eye movements during reading by integrating visual and linguistic information.
Project background and motivation: Many children and adults struggle to attain reading proficiency. Contrary to our subjective experience we only see clearly 7 to 10 letters at a time when we read. Because we do not see the text on the page all at once, we move our eyes through the text during reading. Therefore, skilled reading is defined by eye movements that efficiently gather visual information. Eye tracking is the ideal method for gathering information about eye movements during reading. We can watch and measure the process of meaning extraction in real time. The field has large datasets of eye movements but poor models of those eye movements. We do not understand the knowledge and processes that underlie skilled reading. The goal of this project is to use deep learning to develop a better model capable of predicting eye movements during reading by integrating visual and linguistic information. Better model of skilled reading will help inform reading interventions and education by giving more detailed reader profiles.
This project is a collaboration between Learning and Language Lab in the Department of Psychology and the Center for the Artificial Intelligence Innovation at NCSA.
Skill qualifications: The intern will assist in developing a deep learning recurrent network model to predict reading skills of human participants based on their eye movements.
Skills required:
The project centers on DISSCO, software for composition, sound design and music notation/printing developed at UIUC, NCSA and Argonne National Laboratory. Written in C++, it includes a Graphic User Interface using gtkmm. A parallel version has been developed at the San Diego Supercomputer Center with support from XSEDE (Extreme Science and Engineering Discovery Environment).
DISSCO has a directed graph structure and uses stochastic distributions, sieves (part of Number Theory), Markov chains and elements of Information Theory to produce musical compositions. Presently, efforts are directed toward refining a system for the notation of music as well as to the realization of an Evolving Entity, a composition whose aspects change when computed recursively over long periods of time thus mirroring the way living organisms are transformed in time (Artificial Life).
Another possible direction of research is sonification of complex scientific data, the aural rendition of computer generated data, as a companion to visualization.
More information about:
Skills required:
Preferred skills:
Neutron stars are extraordinary not only because they are the densest form of matter in the visible Universe but also because they have magnetic fields which can reach levels that can distort the very nature of quantum vacuum. In this project, with the help of supercomputers we will study the combined gravitational and electromagnetic field of a neutron star in a self-consistent way in order to create a realistic model for the first time.
The successful applicant will use the Einstein Toolkit to perform astrophysical simulations of magnetized neutron stars that will help understand better multimessenger events like GW170817. Theoretical work in magnetohydrodynamics will also be possible.
Early Detection and Prediction of Parkinsonism Powered by Multi-Modal Few-Shot Learning
Yuxiong Wang
In this project, we focus on discriminating several indicators that are associated with Parkinsonism, such as slurred speech, asymmetric facial expression, stooped posture, abnormal gait, and tremor. We will utilize the most recent, powerful self-supervised deep learning methods and further extend them into the cross-modality scenario to learn discriminative feature representations for data from different modalities (ranging from images, videos, and audios) with the small-size annotated dataset. The project is looking for a student familiar with Computer Vision, Deep Learning, Python, and PyTorch.
