SPIN Projects for AY 2023-24

	B	C	D	H
1	Mentor Name	Project Title	Project Description	Comments
2	Bruno Abreu	Neutral-atom simulations as a platform for distributed quantum computing interfaces	Analog quantum computing with neutral-atom systems constitutes a promising platform for prototyping and experimenting with distributed quantum computing. Their relatively large number of qubits and great level of geometric control provided by optical tweezers allows for modeling and testing quantum communication between small clusters of qubits with and without entanglement across the entire system. This flexibility is a significant convenience for developing hybrid quantum-classical state passing interfaces from a programming perspective. In this project, the SPIN intern will simulate analog quantum computers based on the physics of Rydberg atoms using Bloqade. This Julia-based simulator solves the quantum dynamics of the system given protocols for ramping and tuning the parameters of a standard Hamiltonian. The primary goal is to develop protocols to adiabatically create entangled and unentangled clusters of qubits within the limits of current quantum computing technologies in terms of coherence times and lattice sizes. Pre-requisites - Interest in quantum computing applications - Familiarity with basic quantum mechanics and quantum computing concepts - Python and/or Julia programming skills	This is a new project
3	Luke Olson	Center for Exascale-enabled Scramjet Design (CEESD)
4	Shirui Luo	PROMPT-MATCHED SEMANTIC SEGMENTATION	In this project, the intern will work on a system that can generate image segmentations based on arbitrary prompts at test time. The process is also termed “query segmentation”, which draws an analogy to query-based information retrieval systems. Specifically, the original image serves as the “database”, the prompt image or text represent the “key”, and the resulting segmentation represents the queried output. The trained model functions as a query interface, which matches the prompt's description with the entire image and returns relevant raster pixels that correspond to the prompt description.
5	Volodymyr Kindratenko	Investigating the Power of Large Language Models for Workflow Management	In this project, our aim is to investigate the effectiveness of Large Language Models (LLMs) in enhancing our ability to interact with complex systems and workflows. As a student, you will be responsible for 1) Assisting with training/finetuning existing LLMs for specific tasks, 2) Integrating LLMs with cloud and hpc systems, 3) Enabling enhanced functionality. We are seeking candidates who 1) Have experience with NLP models, and 2) Are proficient with a conventional MLOps pipeline.
6	Volodymyr Kindratenko	Developing AI-Based Chatbots for Subject-Specific Tasks	In this project, we aim to develop AI-based chatbots that can quickly be trained on subject-specific corpuses to act as assistants. As a student, your responsibilities will include: 1) Developing, training/finetuning, and implementing NLP models, 2) Integrating NLP models with a chatbot interface . We are looking for candidates who: 1) Have some experience with NLP and chatbot technologies, 2) Possess strong programming skills in Python, 3) Know how to build interactive websites using modern web technology.
7	Ruby Mendenhall	STEM IL Nobel Project Research Team	The STEM IL Nobel Project is a program designed to provide unprecedented access to STEM experiences and technologies to marginalized student populations. In this project, the SPIN intern will assist the research team in program related activities including data analysis, website development, research write-ups, and/or other activities depending on the needs of the program. Pre-requesites: - Interest in furthering diversity, equity, and inclusion in STEM fields - Ability to take initiative on projects - Prior coding and research experience is preferred, but not required
8	Anastasia Stoops	Understanding eye movements in skilled readers and novice readers: A behavioral and computational approach	In this project we apply deep learning algorythms to predict reading skill from eye movements recorded during naturalistic reading of multiline texts. The intern will work under the guidance of the project team members on optimizing existing neural nets and/or creating new ones to simulate human eye movement behavior based on visual and linguistic text features. We are looking for students who: 1. are interested in understanding reading development; 2. have familiarity with data analyses and visualization; 3. have demonstratable prior experience with PyTorch (show completed projects)
9	Becky Smith	RShiny apps for vector control	In this project, we are developing free online tools to assist vector control agencies in their work. This will include converting existing R analysis scripts to RShiny apps, and developing user-friendly interfaces for input and interpretation. This work is sponsored by the CDC, who may add successful apps to their website.
10	Qiyue Lu	Integration schemes over different type and order finite elements	The goal is to develop a multi-physics code based on the finite element method. In the past two three years, the finite element model for heat conduction problem and for laminar incompressible flow have been developed. In this process, it is noticed that the integration schemes (then performance) vary according to the element type and order, and the integrands. In this project, integration schemes over different type and order finite elements will be explored. The goal is to develop a C/C++ library or Fortran subroutine to handle integration over different type elements, therefore, to pave the way for further multi-physics code development.	It requires the SPIN student has a strong applied mathematics background.
11	Yuxiong Wang	Early Detection and Prediction of Parkinsonism Powered by Multi-Modal Few-Shot Learning	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.
12	Roland Haas	Automated simulation management and monitoring for the Einstein Toolkit	The selected student will work with the NCSA gravity group to develop a Python based framework to automate management of numerical relativity simulations of binary black hole collisions. This project will involve actually handling a number of simulations on compute clusters at NCSA and in the ACCESS compute network, and learning all the analysis steps required to produce data ready for a publication. A set of (Python) codes to monitor the simulation status in real time will be devloped and deployed at NCSA. This will involve some 2D plots using matplotlib or similar plotting functionality, possibly 2D movies using VisIt and extraction of textual and numeric information from ASCII and HDF5 output files. A framework to identify typical error conditons and alert users about them will be devloped. Skills required: * Python * matplotlib, numpy, scipy * Linux command line and scripting, including ssh, scp, rsync * basic HTML and very basic CSS skills * working knowledge of git
13	Eliu Huerta	AI surrogates for science	The selected student will work on the development of AI surrogates for one of the following topics: multi-scale and multi-physics simulations in particular turbulence in magnetized fluids; gravitational wave detection of black hole mergers; molecular dynamics simulations of inorganic crystals Skills: Knowledge of TensorFlor or PyTorch, Python, use of supercomputers clusters like Delta or HAL will be a plus
14	Eliu Huerta	Garden AI Models	Description: the selected student will learn to automate the production of AI models, and their use combining scientific data facilities, supercomputers, federated learning and Globus compute. Skills: Knowledge of TensorFlor or PyTorch, Python, use of supercomputers clusters like Delta or HAL will be a plus
15	Angela Lyons, Aiman Soliman	A Machine Learning and Geospatial Approach to Targeting Humanitarian Assistance Among Syrian Refugees in Lebanon	Project Description: 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. Student Contributions:We are looking for a student with experience in basic programming in Python and/or R and basic knowledge and skills in machine learning; experience with GIS and geospatial analysis is a plus. Anticipated tasks include assisting the team with: (1) data preprocessing, (2) data modeling, analysis, and predictions, and (3) the creation of mappings and other data visualizations. The student will develop and review code and create documentation for the code. They will also assist in developing machine learning algorithms and then training, validating, and testing the algorithms. The student will also create a GITHUB for the team, where they will prepare and upload scripts and other documentation for the project. The student will meet with the mentors on a regular basis, participate in team meetings, and actively engage with graduate students. Preferred Skills: • Background in data science and statistical modeling • Programming languages: Python, R, and/or Stata • Basic knowledge and skills in machine learning and/or geospatial analysis • Expertise in creating mappings and other data visualizations • Experience in programming and development of dashboards
16	Yifang Zhang	Data Acquisition for AI Applications at Molecule Maker Lab Institute	Project Description: The Molecule Maker Lab Institute (MMLI) is an interdisciplinary initiative with leaders in AI and organic synthesis intensively collaborating to create frontier AI tools, dynamic open access databases, and fast and broadly accessible small molecule manufacturing and discovery platforms. Advanced AI and machine learning (ML) methods enable the MMLI to achieve AI-enabled synthesis planning, catalyst development, molecule manufacturing, and molecule discovery. In this particular project, student will expect to colleberatively develop a data ingestor tool with the MMLI-devop team to take multiple different form of chemical datasets into an existing database using the CLI (Command-Line Interface) in Python. The main challenge and innovation will be how to design all those chemistry data to fit a relational database using all different kind of parsing strategies. Student Contributions: We are looking for a student with experience in advanced programming in Python. Anticipated tasks include assisting the team with: • Preprocess Chemistry data. • Implement Python code on the application via Github. • Code review with MMLI team member(s). • Document the features onto Github Repo. • (Optional) Design and implement the front-end for the application. Preferred Skills: • Background in CS or ECE related major. • Programming language: Python. • Experience in raw data preprocess. • Experience in Objective Oriented Programming and Github Repository. • Familiar with Chemistry related topics. • Experience with application deployment.
17	Kevin Chang	Academic Online -- Building a Living Encyclopedia for an Academic Domain	To help people learn knowledge of academic topics in a data-driven, AI-built encyclopedia connected with an academic domain (such as Computer Science). Develop algorithms and models to– Automatically discover keywords (concepts or entities, e.g., “data structure”, “neural network”) used in an academic domain (e.g., computer science). Build an “encyclopedia” for these keywords. Organize knowledge, literature, and scholars in the academic domain by these keywords. Techniques: natural language processing, machine learning, data mining.
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100