Current Projects

Angela Lyons, Professor, College of Agricultural, Consumer and Environmental Sciences, Agricultural & Consumer Economics
Aiman Soliman, Research Assistant Professor, College of Fine & Applied Arts, Department of Regional and Urban Planning, Research Scientist at the National Center for Supercomputing Applications (NCSA)

Project Description: 

This interdisciplinary research project examines how internal migration in Northern Brazil and the Amazon—driven by environmental degradation and economic hardship—is reshaping land use, regional economies, and ecological sustainability. Rapid deforestation, fueled by land speculation, agricultural expansion, and poorly designed policies, has disrupted ecosystems, accelerated climate change, and created harsh living conditions that push vulnerable populations to migrate. These migration flows often move people to equally fragile areas, potentially perpetuating cycles of deforestation

Using AI-based geospatial and machine learning methods, our team investigates:
• Where people are moving
• How these movements affect local economies and land use
• The role of policies, credit access, and environmental risk management in shaping migration dynamics

The ultimate goal is to inform sustainable development strategies that balance economic needs with conservation efforts in the Amazon and similar forested regions worldwide.

Student Contributions: 

Responsibilities:
The student will work closely with NCSA researchers, faculty, and graduate students to:
1. Preprocess geospatial and socioeconomic datasets
2. Conduct data modeling, analysis, and predictions
3. Create maps and other data visualizations from geospatial and socioeconomic data
4. Develop, review, and document code in Python and/or R
5. Assist with the development, training, validation, and testing of machine learning algorithms
6. Create and maintain a GitHub repository for scripts and documentation
7. Participate in regular mentor meetings and team discussions

Preferred Qualifications:
• Undergraduate student in data science, computer science, electrical and computer engineering, statistics, or related field
• Proficiency in Python and/or R
• Basic knowledge of machine learning techniques and/or geospatial analysis
• Experience with GIS tools (e.g., QGIS, ArcGIS) and/or remote sensory data is a plus
• Ability to create maps, dashboards, and visual analytics
• Familiarity with version control systems (e.g., Git/GitHub)
• Strong problem-solving skills and attention to detail

Eliu Huerta, College of Liberal Arts and Sciences, Department of Astronomy
Hao Peng, The Grainger College of Engineering, Siebel School of Computing

Project Description: 

The selected student will participate in the development and evaluation of generative AI and scientific large language models for the in silico discovery of materials for energy storage and conversion, including metal organic frameworks and crystal-like materials.

Student Contributions: 

Preferred Qualifications:

Hands-on knowledge using computational chemistry software, such as LAMMPS, DFT and/or GCMC will be a plus. Students should have experience using python, and popular AI APIs (PyTorch, TensorFlow, etc.,). Experience using high performance computing platforms will be a plus.

Volodymyr Kindratenko, Assistant Director at the National Center for Supercomputing Applications (NCSA), Director for the Center for Artificial Intelligence Innovation (CAII), Adjunct Associate Professor in Electrical and Computer Engineering (ECE), Research Associate Professor in Computer Science (CS)
Sharon Donovan, College of Agricultural, Consumer and Environmental Sciences Nutritional Sciences, Professor and Melissa M. Noel Endowed Chair in Nutrition and Health; Director of the Personalized Nutrition Initiative

Project Description: 

This research project aims to develop an AI-powered mobile application that provides both qualitative and quantitative nutrition counseling. Users will log their meals by uploading photos, which are analyzed by a computer vision system to categorize foods and estimate their nutritional profiles. A large language model (LLM)-based chatbot then uses this information—along with the user’s health goals, relevant personal data, and academic nutrition literature—to deliver personalized dietary recommendations. By leveraging generative AI models, the application seeks to make healthy eating guidance more accessible and actionable, helping users adopt sustainable diet changes.

Student Contributions: 

Students will play an active role in designing and developing a user-friendly mobile application dedicated to collecting and analyzing nutritional data. Their work will include testing and validating the application to ensure accuracy and usability. In addition, students will contribute to building the data analytics pipeline that processes nutrition data and generates personalized dietary recommendations, integrating computer vision and AI-driven insights into the overall system.

Yaoyao Liu, Assistant Professor School of Information Sciences and Coordinated Science Laboratory, affiliated with Siebel School of Computing and Data Science and the Department of Electrical & Computer Engineering
Yingying Li, Assistant Professor The Grainger College of Engineering, Industrial & Enterprise Systems Engineering

Project Description: 

Modern computer vision systems increasingly operate in dynamic environments where new object categories and data distributions emerge over time. However, most large pre-trained vision models are trained under static assumptions and struggle to adapt incrementally without catastrophic forgetting. This project will develop a unified class-incremental learning framework that jointly integrates adaptive learning algorithms, data-centric policies, and flexible model architectures. The system will leverage online learning and reinforcement learning to dynamically control data filtering, replay, and ordering for efficient learning from non-stationary data streams. In addition, 3D-aware generative models will be incorporated to improve robustness to occlusions, viewpoint changes, and complex visual structures.

Student Contributions:

The student will assist in implementing and training class-incremental learning models using PyTorch and large-scale vision datasets. They will support data preprocessing, experimentation with adaptive data replay and filtering strategies, and evaluation of model performance under continual learning settings. The student will collaborate closely with graduate students and faculty mentors, document progress through code and reports, and present results in regular research meetings.

Dr. Zhi-Pei Liang, Electrical and Computer Engineering 
Dr. Yudu Li, Bioengineering 

Project Description: 

Brain mapping is one of the most exciting frontiers of contemporary science, offering unprecedented opportunities to enhance our understanding of brain function and develop treatments for brain diseases. With the rapid development of AI technologies, a new era is dawning for brain mapping technology development and applications that promise to transform our understanding of the brain and further revolutionize healthcare. This project aims to develop physics- and biology-driven generative AI models to capture the complex anatomical, functional, and molecular distributions from open-access multimodal brain MRI datasets. These models will be applied to automated brain lesion detection and segmentation (e.g., tumors and stroke), with integrated uncertainty quantification capability. This project offers a unique opportunity for students to contribute to cutting-edge research at the intersection of AI, medical imaging, and neuroscience. 

Student Contributions: 

The REU student will gain hands-on experience in multimodal MRI image pre-processing, implementation of state-of-the-art generative AI models using open-source frameworks (e.g., PyTorch and TensorFlow), as well as model training, validation, and testing on GPU-based supercomputers at NCSA. These skills will enable the student to make meaningful contributions to the development and evaluation of advanced AI-driven brain mapping tools for the project.  

Aiman Soliman, Research Assistant Professor, College of Fine & Applied Arts, Department of Regional and Urban Planning, Research Scientist at the National Center for Supercomputing Applications (NCSA)
Volodymyr Kindratenko, Assistant Director at the National Center for Supercomputing Applications (NCSA), Director for the Center for Artificial Intelligence Innovation (CAII), Adjunct Associate Professor in Electrical and Computer Engineering (ECE), Research Associate Professor in Computer Science (CS)
Zeynep Madak-Erdogan , Professor of Nutrition; Sylvia D. Stroup Scholar

Project Description: 

We are developing a spatially aware AI framework that enables natural-language retrieval and reasoning over spatial biological data. The project focuses on representing spatial organization as a learning object, rather than treating spatial patterns as visual artifacts or post-hoc statistics.

Building on this representation, we will implement a retrieval-augmented generation (RAG) system that allows users to query spatial omics datasets in natural language and receive grounded, interpretable summaries across cohorts. This project is ideal for students interested in computer vision, spatial informatics, graph representations, and large language models, and offers hands-on experience at the intersection of AI and spatial biology.

Student Contributions:

Students will help build a multi-modal embedding system that jointly represents spatial patterns extracted from images and spatial descriptors expressed in language (e.g., scattered, clustered, overlapping), which captures geometric and topological relationships independent of gene identity, in addition to sample-specific biological metadata, such as co-regulated gene sets. 

Alexandre (Alex) Tartakovsky, Professor, The Grainger College of Engineering, Civil & Environmental Engineering, NCSA staff and faculty affiliate

Project Description: 

Groundwater supplies a large fraction of public, agricultural, and rural water demand in the United States, making reliable estimation and forecasting of groundwater levels critical for sustainable management. This project will develop an AI modeling framework that can be trained and used for inference on sparse, irregular groundwater measurements, leveraging attention-based set pooling without filtering or imputation. Heterogeneous auxiliary groundwater and precipitation data will be integrated using multi-source cross-attention. Large-scale predictions will be enabled via regional latent states coupled via a graph attention network. By treating these latent states as part of a dynamical system, the model supports both spatiotemporal interpolation and long-term, scenario-based groundwater forecasting across multiple aquifers.

Student Contributions:

Student participants will participate in the development and training of the AI model.

Yang Wang, Professor, School of Information Sciences

Yun Huang, Associate Professor, School of Information Sciences

Project Description: 

This project develops a multi‑modal AI safety framework that ensures safe, age‑appropriate AI interactions for youth across text, image, audio, and video modalities. It advances AI safety beyond simple content filtering by integrating model‑level guardrails, adaptive interaction policies, and rigorous youth‑focused benchmarks. The work will produce a multi‑modal safety benchmark, a cross‑modal guardrail model, and context‑aware safety policies that adapt to age, task, and interaction mode. It will also introduce transparent safety signals and evaluation protocols that allow safety mechanisms to be measured, compared, and audited. Overall, the project aims to establish a reproducible, trustworthy foundation for youth‑centered AI safety across diverse forms of AI interaction.

Student Contributions:

The student researcher will help build multi‑modal safety benchmark datasets and define youth‑specific risk categories that expand our current text‑focused taxonomy. They will prototype a guardrail model capable of detecting risks across text, images, audio, and video, and implement evaluation pipelines to test these interactions. Their work will compare traditional rule‑based filters with learned guardrail approaches to understand trade‑offs in safety, usefulness, and user autonomy. The student will analyze results, document benchmark performance, and translate findings into system design recommendations. Through this work, they will contribute to both technical AI safety research and human‑centered design for youth‑appropriate AI systems.

Anastasia Stoops, Research Scientist, Psychology Department

Vlad Kindratenko, Research Associate Professor, National Center for Supercomputing Applications

Project Description: 

The main goal of this project is to understand through a process known as “AI-powered digital phenotyping” how eye-movement profiles change as a function of the reader’s experience with a text’s visual and linguistic features. Many children and adults struggle to attain reading proficiency. Contrary to our subjective experience, readers perceive only 7–10 letters at a time, which requires continuous eye movements to extract meaning from text. Skilled reading is therefore characterized by efficient eye movement patterns that support the uptake of visual and linguistic information. The project will use validated Provo-ILLINI corpus (Luke & Christianson, 2018; Stoops et al., in prep). Currently this is the largest open-source database that contains eye movements recorded from adolescents and young adults while they read multi-line connected texts and supplemented with reading expertise assessments for each participant. We aim to fine-tune pre-trained deep learning transformer on text features and reading expertise assessments to predict individual eye-movement profiles. Overall, the project aims to develop an AI-powered computational model that predicts and diagnose readers skill and/or deficiencies from eye-movements and readers experience with text related visual and linguistic features.

Student Contributions:

The student researcher will help build deep learning transformer architecture that allows for multi-level modeling of the readers eye movement profiles. Taking a step-by-step approach, the student researcher will create three instances of fine-tuned pre-trained transformers: 1) fine-tuned on text features; 2) fine-tuned on readers’ language expertise assessments; 3) fine-tuned on both text and reader related features. Next, student researcher under the guidance of the faculty mentors will assess the accuracy of the model generated eye movement patterns and examine the knowledge the models used by extracting information-theoretic benchmark assessments across different layers within the architecture for each instance of the transformer.