This project explores how artificial intelligence (AI) can be used to make satellite observations of air pollution more accessible, interactive, and insightful. Students will design an AI agent capable of automatically retrieving and visualizing satellite-based observations of air quality, such as nitrogen dioxide (NO2), from geostationary and polar-orbiting sensors (e.g., NASA’s TEMPO and ESA’s TROPOMI missions). The AI agent will interpret natural-language commands from users (e.g., “Show today’s (...)
RAD Collaboratory SURF
SEBS - Environmental Science
SEED2S- DS Approved, Artificial Intelligence, Data (...)
"Immune cells eliminate tumors through tightly regulated interactions with cancer cells, yet the molecular and structural mechanisms guiding this process remain poorly understood. When an immune cell contacts a tumor cell, they form an immunological synapse—a dynamic interface where spatial organization and signaling determine whether the tumor is destroyed or evades immune attack. Each participating cell also carries a unique gene expression signature that encodes its functional state and responsiveness. (...)
RAD Collaboratory SURF
SAS - Computer Science
Addiction, Data Science
"Recent algal blooms of Sargassum are causing an environmental disaster since 2011 in the waters and coasts of Puerto Rico and the Caribbean, more recently in Florida, and likely for the same to occur in other Gulf states in the near future. This has caused a severe crisis in the tourism industry of the region, the local economies, and their infrastructure to manage waste, overwhelming landfills and incinerators. Therefore, there is a pressing need to find solutions to utilize this unused feedstock (...)
RAD Collaboratory SURF
SAS - Chemistry & Chemical Biology
Artificial Intelligence, Data Science
This project aims to advance in silico drug discovery using state-of-the-art artificial intelligence (AI) models. The work involves four key components. First, we will develop an accurate and scalable docking model to describe protein–drug binding. Second, an autoregressive model capable of generating molecular structures with 3D atomic precision will be integrated with the docking model. Third, large-scale training will be carried out using publicly available protein–drug docking data and refined (...)
RAD Collaboratory SURF
SAS - Chemistry & Chemical Biology
Artificial Intelligence, Data Science
Deep Interaction Prediction Network (DIPN, https://arxiv.org/pdf/2011.04692) demonstrates impressive capability in predicting multi-object dynamics during robotic pushing, using modular MLPs to model both direct and interactive transformations among objects. However, its MLP-based interaction modules treat object-object relations in a pairwise and static manner, limiting scalability and expressiveness when complex, long-range dependencies occur. This project proposes to replace DIPN’s MLP interaction (...)
RAD Collaboratory SURF
SAS - Computer Science
Artificial Intelligence
This project aims to perform a literature review on control theory applications to LLMs and develop preliminary implementations of relevant control methods to potentially influence their behavior.
RAD Collaboratory SURF
Engn - Mechanical & Aerospace Engineering
Artificial Intelligence
"The objective of this project is to create a library of trained diffusion models that generate conditional time series for use in energy system models. Examples include “a 24-hour injection pattern of an offshore wind farm at the NJ coast in spring” or “one week of power consumption of a multi-family residential building in summer.”
Energy system models are essential decision-support tools in engineering, finance, and policymaking. Emerging weather- and behavior-driven energy resources (wind, (...)
RAD Collaboratory SURF
Engn - Industrial Engineering
Data Science
Finite-element-based physics-informed neural networks (FE-PINNs) provide a strategy for training computationally efficient and flexible surrogate models. Existing FE-PINN architectures only feature custom convolutional operators, which greatly limits architecture design. In this project, the student will implement new custom operators that enable encoder-decoder style networks (e.g., U-Net). These custom operators will be incorporated into the existing software package for FE-PINN training.
RAD Collaboratory SURF
Engn - Materials Science and Engineering
Artificial Intelligence, Data Science
"RCSB Protein Data Bank (RCSB PDB, http://rcsb.org) is a global online resource that provides access to atomic level information about proteins, nucleic acids, and complex macromolecular assemblies available in the PDB archive through development of tools and resources for research and education in molecular biology, structural biology, computational biology, and beyond.
PDB data are crucial to users around the world; our website supports many millions of users each year. PDB data are also redistributed (...)
RAD Collaboratory SURF
New Brunswick Office for Research
Artificial Intelligence, Data Science
"We propose to develop an open-source agentic science assistant that helps students and researchers explore, visualize, and interpret large, heterogeneous datasets focused on the ocean. Approximately half of the data the National Oceanic and Atmospheric Administration (NOAA) has collected (32 petabytes) is hosted by Amazon Web Services [1]. The Rutgers University Center for Ocean Observing Leadership (RUCOOL) and other geoscience researchers (there are approximately 300,000 in the United States) (...)
RAD Collaboratory SURF
SEBS - Marine & Coastal Sciences
Artificial Intelligence, Data Science
The current project aims to improve spectrophotometric testing—a branch of chemical physics—in environmental applications by applying AI/ML techniques. Spectrophotometry is a powerful analytical technique based on the principle that every chemical absorbs light at specific wavelengths, enabling identification and quantification of hundreds of substances in chemistry, biology, and medicine (think color-change-based measurements in blood glucose monitors, chlorine test strips in swimming pools, or (...)
RAD Collaboratory SURF
SEBS - Environmental Science
Computer Science and Engineering, Information Technology, (...)
We will explore several new optimization approaches to accelerate the pretraining of large language models (LLMs). Candidate approaches include second-order optimization and gradient orthogonalization. The student will examine the performance of these optimizers on GPT model families. We will also design new approaches to lower the communication cost of these optimizers.
RAD Collaboratory SURF
Engn - Electrical & Computer Engineering
Artificial Intelligence
"Artificial Intelligence (AI) is rapidly transforming education by offering new ways to support student learning and success. This project will examine how AI tools can enhance academic performance, engagement, and overall well-being in undergraduate Economics and Business courses. Students will work with me to analyze how AI-based tools, such as personalized learning platforms, AI chatbots, and feedback systems—can create individualized learning pathways. These tools help educators deliver timely (...)
RAD Collaboratory SURF
SEBS - Agricultural & Resource Management Agents
Data Science
"Have you ever wondered how a robot learns to clean a room, sort recycling, or organize a warehouse? This project tackles these challenges by building a robotic system that can see its environment, decide what to do, and learn how to act. You will be part of a team that integrates cameras and robotic arms. The robot will start with a pile of jumbled objects on a table (e.g., soda cans, water bottles, and coffee mugs). Using its camera and a natural language instruction, it must identify the target (...)
RAD Collaboratory SURF
SAS - Computer Science
Artificial Intelligence
