I2S Masters/ Doctoral Theses


All students and faculty are welcome to attend the final defense of I2S graduate students completing their M.S. or Ph.D. degrees. Defense notices for M.S./Ph.D. presentations for this year and several previous years are listed below in reverse chronological order.

Students who are nearing the completion of their M.S./Ph.D. research should schedule their final defenses through the EECS graduate office at least THREE WEEKS PRIOR to their presentation date so that there is time to complete the degree requirements check, and post the presentation announcement online.

Upcoming Defense Notices

Arya Hadizadeh Moghaddam

Learning Personalized and Robust Patient Representations across Complex Structures in Electronic Health Records

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Dissertation Defense

Committee Members:

Zijun Yao, Chair
Fengjun Li
Bo Luo
Han Wang
Xinmai Yang

Abstract

Electronic health records (EHRs) contain medical codes and clinical notes collected over multiple visits. This dissertation studies how to learn patient representations from these records through three related aspects problems:

The first aspect concerns relationships within a patient’s history. TPGT models changing dependencies among medical codes within and across visits, and ARCI separates temporal patterns associated with different medical intents. They account for interacting clinical events and concurrent treatment goals.

When patient histories are limited, these patterns are harder to learn. The second aspect is addressed by MetaDrug, which adapts to patients with sparse histories using their own records and those of similar patients, while filtering uncertain samples.

The third aspect concerns incorporating clinical notes and large language models into patient representations. RePrompT uses cohort-trained representations and recurrent soft prompts to preserve longitudinal context. RASPER trains a note summarizer through downstream prediction feedback. CERE reasons over notes conditioned on structured codes to represent the additional evidence they provide. These studies connect structured histories with clinical text to support prediction.

Experiments on the KUMC acute kidney injury cohort, eICU, MIMIC-III, and MIMIC-IV show improvements over the respective baselines in clinical prediction and medication recommendation.


Shelby Jones

Measuring LLM Susceptibility to Stylistic Perturbations in Social Media Text

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

MS Thesis Defense

Committee Members:

Jennifer Lohoefener, Chair
Perry Alexander
John Symons


Abstract

Large Language Models (LLMs) are becoming increasingly integrated into important domains such as social media and politics. This makes it essential that we understand their susceptibility to superficial text variations. In this thesis, I investigated whether small stylistic perturbations in social media posts influence how an LLM perceives a post's political leanings. To do this, I systematically manipulated punctuation and capitalization across both political and apolitical social media posts. I then used VADER sentiment analysis to measure the changes in semantic intensity to quantify the text manipulation. Finally, I prompted each LLM to provide its perceived political leanings of each given post. I used linear regression to assess the VADER score correlation with the LLMs' political alignment ratings. While there was no statistically significant relation between perceived political extremity and VADER scores, category-level analysis revealed that the stylistic perturbations cause significant inconsistencies in model outputs. Furthermore, in posts with neutral semantic content, the stylistic perturbations introduce a right-leaning bias. These findings add to the growing body of literature on LLM instability under prompt perturbation and highlight risks associated with automated content analysis. 


Connor Sullivan

Memory QoS Mechanisms for Predictable and Performant Real-Time Systems

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Comprehensive Defense

Committee Members:

Heechul Yun, Chair
Zitao Chen
Tamzidul Hoque
Prasad Kulkarni
Kenneth Fischer

Abstract

Commercial-off-the-shelf (COTS) multicore processors are increasingly used in safety-critical real-time systems due to their higher performance and improved size, weight, and power (SWaP) characteristics. However, modern platforms are designed for average-case performance, which works in direct opposition to the worst-case timing guarantees required by safety-critical systems.

In particular, unregulated sharing of the memory hierarchy across cores can cause severe and unpredictable interference. When multiple cores concurrently access shared memory resources, their requests contend with one another, degrading application performance and undermining timing guarantees. Hardware quality-of-service (QoS) mechanisms aim to manage this contention. Partitioning these resources in space alone does not guarantee temporal isolation. Bandwidth regulation can protect latency-critical tasks from such interference, but it trades system throughput for isolation. Current solutions throttle co-running tasks far more than necessary, offering system designers a poor isolation-throughput tradeoff.

Addressing this challenge, this thesis focuses on QoS mechanisms that improve the balance between throughput and isolation. Shared memory resources such as the last-level cache (LLC) and main memory (DRAM) have parallel, banked structures, yet existing regulation treats them as monolithic. This thesis shows that contention in both arises at the bank level and proposes per-bank bandwidth regulation for the LLC and DRAM, which preserves isolation for latency-critical tasks while recovering throughput for best-effort tasks. Finally, the thesis aims to coordinate regulation across the levels of the memory hierarchy, adapting it dynamically at runtime. It will also explore matching the control mechanism to the root cause of contention, for example by controlling the number of outstanding requests, as an alternative or cooperative means to bandwidth regulation. Together, these efforts work toward a better balance between system throughput and isolation guarantees in multicore real-time systems.


Fairuz Shadmani Shishir

Efficient and Privacy-Preserving Biomedical AI: From Precision Therapeutics Design to Private Clinical Prediction

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Dissertation Defense

Committee Members:

Sumaiya Shomaji, Chair
Tamzidul Hoque
Cuncong Zhong
Bishnu Sarker
Michael Hageman

Abstract

Accurate protein characterization, precision therapeutics design, and privacy-preserving clinical modeling are important challenges in biomedical artificial intelligence (AI). Addressing these challenges requires computational frameworks that are not only accurate and scalable but also capable of supporting reliable decision-making. This dissertation develops efficient, scalable, and reliable computational frameworks across three areas: Computational characterization of proteins, precision therapeutics design, and privacy-aware clinical representation learning. The first part of this dissertation focuses on efficient protein characterization at both the sequence and residue levels. GPCR-SLM, a knowledge-distilled small protein language model, classifies G-protein-coupled receptors (GPCRs), a family targeted by approximately 35% of FDA-approved drugs. Across 86 families, GPCR-SLM achieves 99% accuracy, outperforming BLAST (86.4%) and HMMER (91%), with a 33.5-fold speedup over large protein language models. Extending characterization to the residue level, a protein-metal-ion binding site prediction framework evaluates five protein language models and incorporates positional encoding to capture long-range dependencies. Under 10-fold cross-validation, it achieves a Matthews correlation coefficient of 0.89, with precision, recall, and F1 scores exceeding 95% across six metal ions.

Building on these protein characterization methods, the second part advances precision therapeutics design through uncertainty quantification and generative learning for therapeutic antibodies. NbBayesLM combines protein language model embeddings and physicochemical descriptors in a Bayesian neural network to predict nanobody thermostability. Evaluated on 10,630 VHH sequences, it achieves a mean absolute error of 1.89C and an R2 of 0.67 while estimating epistemic and aleatoric uncertainty to support candidate prioritization. CIgFlow employs classifier-free guidance to generate de novo, antigen-conditioned antibody sequences without fixed scaffolds or paired structural data. Generated sequences achieve a median OASis humanness score of 0.90 and mean predicted local distance difference test (pLDDT) score of 85.0, demonstrating promising novelty and diversity. The third part addresses privacy-aware clinical modeling using electrocardiograms (ECGs), which encode diagnostic information and sensitive demographic attributes. A variational autoencoder with dual discriminators suppresses demographic information while preserving clinically relevant representations.

Collectively, this dissertation advances efficient protein representation learning, computational antibody design, and privacy-preserving clinical AI, contributing methodological foundations for trustworthy biomedical applications.


Past Defense Notices

Dates

Arnab Mukherjee

Attention-Based Solutions for Occlusion Challenges in Person Tracking

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Comprehensive Defense

Committee Members:

Prasad Kulkami, Chair
Sumaiya Shomaji
Hongyang Sun
Jian Li

Abstract

Person tracking and association is a complex task in computer vision applications. Even with a powerful detector, a highly accurate association algorithm is necessary to match and track the correct person across all frames. This method has numerous applications in surveillance, and its complexity increases with the number of detected objects and their movements across frames. A significant challenge in person tracking is occlusion, which occurs when an individual being tracked is partially or fully blocked by another object or person. This can make it difficult for the tracking system to maintain the identity of the individual and track them effectively.

In this research, we propose a solution to the occlusion problem by utilizing an occlusion-aware spatial attention transformer. We have divided the entire tracking association process into two scenarios: occlusion and no-occlusion. When a detected person with a specific ID suddenly disappears from a frame for a certain period, we employ advanced methods such as Detector Integration and Pose Estimation to ensure the correct association. Additionally, we implement a spatial attention transformer to differentiate these occluded detections, transform them, and then match them with the correct individual using the Cosine Similarity Metric.

The features extracted from the attention transformer provide a robust baseline for detecting people, enhancing the algorithms adaptability and addressing key challenges associated with existing approaches. This improved method reduces the number of misidentifications and instances of ID switching while also enhancing tracking accuracy and precision.


Agraj Magotra

Data-Driven Insights into Sustainability: An Artificial Intelligence (AI) Powered Analysis of ESG Practices in the Textile and Apparel Industry

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Project Defense

Committee Members:

Sumaiya Shomaji, Chair
Prasad Kulkarni
Zijun Yao


Abstract

The global textile and apparel (T&A) industry is under growing scrutiny for its substantial environmental and social impact, producing 92 million tons of waste annually and contributing to 20% of global water pollution. In Bangladesh, one of the world's largest apparel exporters, the integration of Environmental, Social, and Governance (ESG) practices is critical to meet international sustainability standards and maintain global competitiveness. This master's study leverages Artificial Intelligence (AI) and Machine Learning (ML) methodologies to comprehensively analyze unstructured corporate data related to ESG practices among LEED-certified Bangladeshi T&A factories.

Our study employs advanced techniques, including Web Scraping, Natural Language Processing (NLP), and Topic Modeling, to extract and analyze sustainability-related information from factory websites. We develop a robust ML framework that utilizes Non-Negative Matrix Factorization (NMF) for topic extraction and a Random Forest classifier for ESG category prediction, achieving an 86% classification accuracy. The study uncovers four key ESG themes: Environmental Sustainability, Social : Workplace Safety and Compliance, Social: Education and Community Programs, and Governance. The analysis reveals that 46% of factories prioritize environmental initiatives, such as energy conservation and waste management, while 44% emphasize social aspects, including workplace safety and education. Governance practices are significantly underrepresented, with only 10% of companies addressing ethical governance, healthcare provisions and employee welfare.

To deepen our understanding of the ESG themes, we conducted a Centrality Analysis to identify the most influential keywords within each category, using measures such as degree, closeness, and eigenvector centrality. Furthermore, our analysis reveals that higher certification levels, like Platinum, are associated with a more balanced emphasis on environmental, social, and governance practices, while lower levels focus primarily on environmental efforts. These insights highlight key areas where the industry can improve and inform targeted strategies for enhancing ESG practices. Overall, this ML framework provides a data-driven, scalable approach for analyzing unstructured corporate data and promoting sustainability in Bangladesh’s T&A sector, offering actionable recommendations for industry stakeholders, policymakers, and global brands committed to responsible sourcing.


Samyoga Bhattarai

Pro-ID: A Secure Face Recognition System using Locality Sensitive Hashing to Protect Human ID

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Project Defense

Committee Members:

Sumaiya Shomaji, Chair
Tamzidul Hoque
Hongyang Sun


Abstract

Face recognition systems are widely used in various applications, from mobile banking apps to personal smartphones. However, these systems often store biometric templates in raw form, posing significant security and privacy risks. Pro-ID addresses this vulnerability by incorporating SimHash, an algorithm of Locality Sensitive Hashing (LSH), to create secure and irreversible hash codes of facial feature vectors. Unlike traditional methods that leave raw data exposed to potential breaches, SimHash transforms the feature space into high-dimensional hash codes, safeguarding user identity while preserving system functionality. 

The proposed system creates a balance between two aspects: security and the system’s performance. Additionally, the system is designed to resist common attacks, including brute force and template inversion, ensuring that even if the hashed templates are exposed, the original biometric data cannot be reconstructed.  

A key challenge addressed in this project is minimizing the trade-off between security and performance. Extensive evaluations demonstrate that the proposed method maintains competitive accuracy rates comparable to traditional face recognition systems while significantly enhancing security metrics such as irreversibility, unlinkability, and revocability. This innovative approach contributes to advancing the reliability and trustworthiness of biometric systems, providing a secure framework for applications in face recognition systems. 


Shalmoli Ghosh

High-Power Fabry-Perot Quantum-Well Laser Diodes for Application in Multi-Channel Coherent Optical Communication Systems

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

MS Thesis Defense

Committee Members:

Rongqing Hui, Chair
Shannon Blunt
James Stiles


Abstract

Wavelength Division Multiplexing (WDM) is essential for managing rapid network traffic growth in fiber optic systems. Each WDM channel demands a narrow-linewidth, frequency-stabilized laser diode, leading to complexity and increased energy consumption. Multi-wavelength laser sources, generating optical frequency combs (OFC), offer an attractive solution, enabling a single laser diode to provide numerous equally spaced spectral lines for enhanced bandwidth efficiency.

Quantum-dot and quantum-dash OFCs provide phase-synchronized lines with low relative intensity noise (RIN), while Quantum Well (QW) OFCs offer higher power efficiency, but they have higher RIN in the low frequency region of up to 2 GHz. However, both quantum-dot/dash and QW based OFCs, individual spectral lines exhibit high phase noise, limiting coherent detection. Output power levels of these OFCs range between 1-20 mW where the power of each spectral line is typically less than -5 dBm. Due to this requirement, these OFCs require excessive optical amplification, also they possess relatively broad spectral linewidths of each spectral line, due to the inverse relationship between optical power and linewidth as per the Schawlow-Townes formula. This constraint hampers their applicability in coherent detection systems, highlighting a challenge for achieving high-performance optical communication.

In this work, coherent system application of a single-section Quantum-Well Fabry-Perot (FP) laser diode is demonstrated. This laser delivers over 120 mW optical power at the fiber pigtail with a mode spacing of 36.14 GHz. In an experimental setup, 20 spectral lines from a single laser transmitter carry 30 GBaud 16-QAM signals over 78.3 km single-mode fiber, achieving significant data transmission rates. With the potential to support a transmission capacity of 2.15 Tb/s (4.3 Tb/s for dual polarization) per transmitter, including Forward Error Correction (FEC) and maintenance overhead, it offers a promising solution for meeting the escalating demands of modern network traffic efficiently.


TJ Barclay

Proof-Producing Translation from Gallina to CakeML

When & Where:


Nichols Hall, Room 250 (Gemini Room)

Degree Type:

PhD Dissertation Defense

Committee Members:

Perry Alexander, Chair
Alex Bardas
Drew Davidson
Sankha Guria
Eileen Nutting

Abstract

Users of theorem provers often desire to to extract their verified code to a  more efficient, compiled language. Coq's current extraction mechanism provides this facility but does not provide a formal guarantee that the extracted code has the same semantics as the logic it is extracted from. Providing such a guarantee requires a formal semantics for the target code. The CakeML project, plemented in HOL4, provides a formally defined syntax and semantics for a subset of SML and includes a proof-producing translator from higher-order logic to CakeML. We use the CakeML definition to develop certifying extractor to CakeML from Gallina using the translation and proof techniques of the HOL4 CakeML translator. We also address how differences between HOL4 (higher-order logic) and Coq calculus of constructions) effect the implementation details of the Coq translator.


Anissa Khan

Privacy Preserving Biometric Matching

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Thesis Defense

Committee Members:

Perry Alexander, Chair
Prasad Kulkarni
Fengjun Li


Abstract

Biometric matching is a process by which distinct features are used to identify an individual. Doing so privately is important because biometric data, such as fingerprints or facial features, is not something that can be easily changed or updated if put at risk. In this study, we perform a piece of the biometric matching process in a privacy preserving manner by using secure multiparty computation (SMPC). Using SMPC allows the identifying biological data, called a template, to remain stored by the data owner during the matching process. This provides security guarantees to the biological data while it is in use and therefore reduces the chances the data is stolen. In this study, we find that performing biometric matching using SMPC is just as accurate as performing the same match in plaintext.


Bryan Richlinski

Prioritize Program Diversity: Enumerative Synthesis with Entropy Ordering

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

MS Thesis Defense

Committee Members:

Sankha Guria, Chair
Perry Alexander
Drew Davidson
Jennifer Lohoefener

Abstract

Program synthesis is a popular way to create a correct-by-construction program from a user-provided specification. Term enumeration is a leading technique to systematically explore the space of programs by generating terms from a formal grammar. These terms are treated as candidate programs which are tested/verified against the specification for correctness. In order to prioritize candidates more likely to satisfy the specification, enumeration is often ordered by program size or other domain-specific heuristics. However, domain-specific heuristics require expert knowledge, and enumeration by size often leads to terms comprised of frequently repeating symbols that are less likely to satisfy a specification. In this thesis, we build a heuristic that prioritizes term enumeration based on variability of individual symbols in the program, i.e., information entropy of the program. We use this heuristic to order programs in both top-down and bottom-up enumeration. We evaluated our work on a subset of the PBE-String track of the 2017 SyGuS competition benchmarks and compared against size-based enumeration. In top-down enumeration, our entropy heuristic shortens runtime in ~56% of cases and tests fewer programs in ~80% before finding a valid solution. For bottom-up enumeration, our entropy heuristic improves the number of enumerated programs in ~41% of cases before finding a valid solution, without improving the runtime. Our findings suggest that using entropy to prioritize program enumeration is a promising step forward for faster program synthesis.


Elizabeth Wyss

A New Frontier for Software Security: Diving Deep into npm

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Comprehensive Defense

Committee Members:

Drew Davidson, Chair
Alex Bardas
Fengjun Li
Bo Luo
J. Walker

Abstract

Open-source package managers (e.g., npm for Node.js) have become an established component of modern software development. Rather than creating applications from scratch, developers may employ modular software dependencies and frameworks--called packages--to serve as building blocks for writing larger applications. Package managers make this process easy. With a simple command line directive, developers are able to quickly fetch and install packages across vast open-source repositories. npm--the largest of such repositories--alone hosts millions of unique packages and serves billions of package downloads each week. 

However, the widespread code sharing resulting from open-source package managers also presents novel security implications. Vulnerable or malicious code hiding deep within package dependency trees can be leveraged downstream to attack both software developers and the users of their applications. This downstream flow of software dependencies--dubbed the software supply chain--is critical to secure.

This research provides a deep dive into the npm-centric software supply chain, exploring various facets and phenomena that impact the security of this software supply chain. Such factors include (i) hidden code clones--which obscure provenance and can stealthily propagate known vulnerabilities, (ii) install-time attacks enabled by unmediated installation scripts, (iii) hard-coded URLs residing in package code, (iv) the impacts open-source development practices, and (v) package compromise via malicious updates. For each facet, tooling is presented to identify and/or mitigate potential security impacts. Ultimately, it is our hope that this research fosters greater awareness, deeper understanding, and further efforts to forge a new frontier for the security of modern software supply chains. 


Yousif Dafalla

Web-Armour: Mitigating Reconnaissance and Vulnerability Scanning with Injecting Scan-Impeding Delays in Web Deployments

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

PhD Dissertation Defense

Committee Members:

Alex Bardas, Chair
Drew Davidson
Fengjun Li
Bo Luo
ZJ Wang

Abstract

Scanning hosts on the internet for vulnerable devices and services is a key step in numerous cyberattacks. Previous work has shown that scanning is a widespread phenomenon on the internet and commonly targets web application/server deployments. Given that automated scanning is a crucial step in many cyberattacks, it would be beneficial to make it more difficult for adversaries to perform such activity.

In this work, we propose Web-Armour, a mitigation approach to adversarial reconnaissance and vulnerability scanning of web deployments. The proposed approach relies on injecting scanning impeding delays to infrequently or rarely used portions of a web deployment. Web-Armour has two goals: First, increase the cost for attackers to perform automated reconnaissance and vulnerability scanning; Second, introduce minimal to negligible performance overhead to benign users of the deployment. We evaluate Web-Armour on live environments, operated by real users, and on different controlled (offline) scenarios. We show that Web-Armour can effectively lead to thwarting reconnaissance and internet-wide scanning.


Daniel Herr

Information Theoretic Waveform Design with Application to Physically Realizable Adaptive-on-Transmit Radar

When & Where:


Nichols Hall, Room 129 (Ron Evans Apollo Auditorium)

Degree Type:

PhD Dissertation Defense

Committee Members:

James Stiles, Chair
Christopher Allen
Shannon Blunt
Carl Leuschen
Chris Depcik

Abstract

The fundamental task of a radar system is to utilize the electromagnetic spectrum to sense a scattering environment and generate some estimate from this measurement. This task can be posed as a Bayesian estimation problem of random parameters (the scattering environment) through an imperfect sensor (the radar system). From this viewpoint, metrics such as error covariance and estimator precision (or information) can be leveraged to evaluate and improve the performance of radar systems. Here, physically realizable radar waveforms are designed to maximize the Fisher information (FI) (specifically, a derivative of FI known as marginal Fisher information (MFI)) extracted from a scattering environment thereby minimizing the expected error covariance about an estimation parameter space. This information theoretic framework, along with the high-degree of design flexibility afforded by fully digital transmitter and receiver architectures, creates a high-dimensionality design space for optimizing radar performance.

First, the problem of joint-domain range-Doppler estimation utilizing a pulse-agile radar is posed from an estimation theoretic framework, and the minimum mean square error (MMSE) estimator is shown to suppress the range-sidelobe modulation (RSM) induced by pulse agility which may improve the signal-to-interference-plus-noise ratio (SINR) in signal-limited scenarios. A computationally efficient implementation of the range-Doppler MMSE estimator is developed as a series of range-profile estimation problems, under specific modeling and statistical assumptions. Next, a transformation of the estimation parameterization is introduced which ameliorates the high noise-gain typically associated with traditional MMSE estimation by sacrificing the super-resolution achieved by the MMSE estimator. Then, coordinate descent and gradient descent optimization methods are developed for designing MFI optimal waveforms with respect to either the original or transformed estimation space. These MFI optimal waveforms are extended to provide pulse-agility, which produces high-dimensionality radar emissions amenable to non-traditional receive processing techniques (such as MMSE estimation). Finally, informationally optimal waveform design and optimal estimation are extended into a cognitive radar concept capable of adaptive and dynamic sensing. The efficacy of the MFI waveform design and MMSE estimation are demonstrated via open-air hardware experimentation where their performance is compared against traditional techniques.