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

David Felton

Optimization and Evaluation of Physical Complementary Radar Waveforms

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

PhD Comprehensive Defense

Committee Members:

Shannon Blunt, Chair
Rachel Jarvis
Patrick McCormick
James Stiles
Zsolt Talata

Abstract

In high dynamic-range environments, matched-filter radar performance is often sidelobe-limited with correlation error being fundamentally constrained by the TB of the collective emission. To contend with the regulatory necessity of spectral containment, the gradient-based complementary-FM framework was developed to produce complementary sidelobe cancellation (CSC) after coherently combining responses from distinct pulses from within a pulse-agile emission. In contrast to most complementary subsets, which were discovered via brute force under the notion of phase-coding, these comp-FM waveform subsets achieve CSC while preserving hardware compatibility since they are FM. Although comp-FM addressed a primary limitation of complementary signals (i.e., hardware distortion), CSC hinges on the exact reconstruction of autocorrelation terms to suppress sidelobes, from which optimality is broken for Doppler shifted signals. This work introduces a Doppler-generalized comp-FM (DG-comp-FM) framework that extends the cancellation condition to account for the anticipated unambiguous Doppler span after post-summing. While this framework is developed for use within a combine-before-Doppler processing manner, it can likewise be employed to design an entire coherent processing interval (CPI) to minimize range-sidelobe modulation (RSM) within the radar point-spread-function (PSF), thereby introducing the potential for cognitive operation if sufficient scattering knowledge is available a-priori. 

Some radar systems operate with multiple emitters, as in the case of Multiple-input-multiple-output (MIMO) radar. Whereas a single emitter must contend with the self-inflicted autocorrelation sidelobes, MIMO systems must likewise contend with the cross-correlation with coincident (in time and spectrum) emissions from other emitters. As such, the determination of "orthogonal waveforms" comprises a large portion of research within the MIMO space, with a small majority now recognizing that true orthogonality is not possible for band-limited signals (albeit, with the exclusion of TDMA). The notion of complementary-FM is proposed for exploration within a MIMO context, whereby coherently combining responses can achieve CSC as well as cross-correlation cancellation for a wide Doppler space. By effectively minimizing cross-correlation terms, this enables improved channel separation on receive as well as improved estimation capability due to reduced correlation error. Proposal items include further exploration/characterization of the space, incorporating an explicit spectral.


Jigyas Sharma

SEDPD: Sampling-Enhanced Differentially Private Defense against Backdoor Poisoning Attacks of Image Classification

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

MS Thesis Defense

Committee Members:

Han Wang, Chair
Drew Davidson
Dongjie Wang


Abstract

Recent advancements in explainable artificial intelligence (XAI) have brought significant transparency to machine learning by providing interpretable explanations alongside model predictions. However, this transparency has also introduced vulnerabilities, enhancing adversaries’ ability for the model decision processes through explanation-guided attacks. In this paper, we propose a robust, model-agnostic defense framework to mitigate these vulnerabilities by explanations while preserving the utility of XAI. Our framework employs a multinomial sampling approach that perturbs explanation values generated by techniques such as SHAP and LIME. These perturbations ensure differential privacy (DP) bounds, disrupting adversarial attempts to embed malicious triggers while maintaining explanation quality for legitimate users. To validate our defense, we introduce a threat model tailored to image classification tasks. By applying our defense framework, we train models with pixel-sampling strategies that integrate DP guarantees, enhancing robustness against backdoor poisoning attacks with XAI. Extensive experiments on widely used datasets, such as CIFAR-10, MNIST, CIFAR-100 and Imagenette, and models, including ConvMixer and ResNet-50, show that our approach effectively mitigates explanation-guided attacks without compromising the accuracy of the model. We also test our defense performance against other backdoor attacks, which shows our defense framework can detect other type backdoor triggers very well. This work highlights the potential of DP in securing XAI systems and ensures safer deployment of machine learning models in real-world applications.


Dimple Galla

Intelligent Application for Cold Email Generation: Business Outreach

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Project Defense

Committee Members:

David Johnson, Chair
Prasad Kulkarni
Dongjie Wang


Abstract

Cold emailing remains an effective strategy for software service companies to improve organizational reach by acquiring clients. Generic emails often fail to get a response.

This project leverages Generative AI to automate the cold email generation. This project is built with the Llama-3.1 model and a Chroma vector database that supports the semantic search of keywords in the job description that matches the project portfolio links of software service companies. The application automatically extracts the technology related job openings for Fortune 500 companies. Users can either select from these extracted job postings or manually enter URL of a job posting, after which the system generates email and sends email upon approval. Advanced techniques like Chain-of-Thought Prompting and Few-Shot Learning were applied to improve the relevance making the email more responsive. This AI driven approach improves engagement and simplifies the business development process for software service companies.


Shahima Kalluvettu Kuzhikkal

Machine Learning Based Predictive Maintenance for Automotive Systems

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Project Defense

Committee Members:

David Johnson, Chair
Rachel Jarvis
Prasad Kulkarni
Hongyang Sun

Abstract

Predictive maintenance plays a central role in reducing vehicle downtime and improving operational efficiency by using data-driven methods to classify the condition of automotive engines. Rather than relying on fixed service schedules or reacting to unexpected breakdowns, this approach leverages machine learning to distinguish between healthy and failed engines based on operational data.

In this project, engine telemetry data capturing key parameters such as engine speed, fuel pressure, and coolant temperature was used to train and evaluate several machine learning models, including logistic regression, random forest, k-nearest neighbors, and a neural network. To further enhance predictive performance, ensemble strategies such as soft voting and stacking were applied. The stacking ensemble, which combines the strengths of multiple classifiers through a meta-learning approach, demonstrated particularly effective results.

This classification-based framework demonstrates how data-driven fault detection can enhance automotive maintenance operations. By identifying engine failures more reliably, machine learning enables safer transportation, reduces maintenance costs, and enhances overall vehicle dependability. Beyond individual vehicles, such approaches have broader applications in fleet management, where proactive decision-making can improve service continuity, reduce operational risks, and increase customer satisfaction.


Jennifer Quirk

Aspects of Doppler-Tolerant Radar Waveforms

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

PhD Comprehensive Defense

Committee Members:

Shannon Blunt, Chair
Patrick McCormick
Charles Mohr
James Stiles
Zsolt Talata

Abstract

The Doppler tolerance of a waveform refers to its behavior when subjected to a fast-time Doppler shift imposed by scattering that involves nonnegligible radial velocity. While previous efforts have established decision-based criteria that lead to a binary judgment of Doppler tolerant or intolerant, it is also useful to establish a measure of the degree of Doppler tolerance. The purpose in doing so is to establish a consistent standard, thereby permitting assessment across different parameterizations, as well as introducing a Doppler “quasi-tolerant” trade-space that can ultimately inform automated/cognitive waveform design in increasingly complex and dynamic radio frequency (RF) environments. 

 

Separately, the application of slow-time coding (STC) to the Doppler-tolerant linear FM (LFM) waveform has been examined for disambiguation of multiple range ambiguities. However, using STC with non-adaptive Doppler processing often results in high Doppler “cross-ambiguity” side lobes that can hinder range disambiguation despite the degree of separability imparted by STC. To enhance this separability, a gradient-based optimization of STC sequences is developed, and a “multi-range” (MR) modification to the reiterative super-resolution (RISR) approach that accounts for the distinct range interval structures from STC is examined. The efficacy of these approaches is demonstrated using open-air measurements. 

 

The proposed work to appear in the final dissertation focuses on the connection between Doppler tolerance and STC. The first proposal includes the development of a gradient-based optimization procedure to generate Doppler quasi-tolerant random FM (RFM) waveforms. Other proposals consider limitations of STC, particularly when processed with MR-RISR. The final proposal introduces an “intrapulse” modification of the STC/LFM structure to achieve enhanced sup pression of range-folded scattering in certain delay/Doppler regions while retaining a degree of Doppler tolerance.


Mary Jeevana Pudota

Assessing Processor Allocation Strategies for Online List Scheduling of Moldable Task Graphs

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Thesis Defense

Committee Members:

Hongyang Sun, Chair
David Johnson
Prasad Kulkarni


Abstract

Scheduling a graph of moldable tasks, where each task can be executed by a varying number of processors with execution time depending on the processor allocation, represents a fundamental problem in high-performance computing (HPC). The online version of the scheduling problem introduces an additional constraint: each task is only discovered when all its predecessors have been completed. A key challenge for this online problem lies in making processor allocation decisions without complete knowledge of the future tasks or dependencies. This uncertainty can lead to inefficient resource utilization and increased overall completion time, or makespan. Recent studies have provided theoretical analysis (i.e., derived competitive ratios) for certain processor allocation algorithms. However, the algorithms’ practical performance remains under-explored, and their reliance on fixed parameter settings may not consistently yield optimal performance across varying workloads. In this thesis, we conduct a comprehensive evaluation of three processor allocation strategies by empirically assessing their performance under widely used speedup models and diverse graph structures. These algorithms are integrated into a List scheduling framework that greedily schedules ready tasks based on the current processor availability. We perform systematic tuning of the algorithms’ parameters and report the best observed makespan together with the corresponding parameter settings. Our findings highlight the critical role of parameter tuning in obtaining optimal makespan performance, regardless of the differences in allocation strategies. The insights gained in this study can guide the deployment of these algorithms in practical runtime systems.


Aidan Schmelzle

Exploration of Human Design with Genetic Algorithms as Artistic Medium for Color Images

When & Where:


Eaton Hall, Room 2001B

Degree Type:

MS Project Defense

Committee Members:

Arvin Agah, Chair
David Johnson
Jennifer Lohoefener


Abstract

Genetic Algorithms (GAs), a subclass of evolutionary algorithms, seek to apply the concept of natural selection to promote the optimization and furtherance of “something” designated by the user. GAs generate a population of chromosomes represented as value strings, score each chromosome with a “fitness function” on a defined set of criteria, and mutate future generations depending on the scores ascribed to each chromosome. In this project, each chromosome is a bitstring representing one canvased color artwork. Artworks are scored with a variety of design fundamentals and user preference. The artworks are then evolved through thousands of generations and the final piece is computationally drawn for analysis. While the rise of gradient-based optimization has resulted in more limited use-cases of GAs, genetic algorithms still have applications in various settings such as hyperparameter tuning, mathematical optimization, reinforcement learning, and black box scenarios. Neural networks are favored presently in image generation due to their pattern recognition and ability to produce new content; however, in cases where a user is seeking to implement their own vision through careful algorithmic refinement, genetic algorithms still find a place in visual computing.


Andrew Riachi

An Investigation Into The Memory Consumption of Web Browsers and A Memory Profiling Tool Using Linux Smaps

When & Where:


Nichols Hall, Room 250 (Gemini Room)

Degree Type:

MS Thesis Defense

Committee Members:

Prasad Kulkami, Chair
Perry Alexander
Drew Davidson
Heechul Yun

Abstract

Web browsers are notorious for consuming large amounts of memory. Yet, they have become the dominant framework for writing GUIs because the web languages are ergonomic for programmers and have a cross-platform reach. These benefits are so enticing that even a large portion of mobile apps, which have to run on resource-constrained devices, are running a web browser under the hood. Therefore, it is important to keep the memory consumption of web browsers as low as practicable.

In this thesis, we investigate the memory consumption of web browsers, in particular, compared to applications written in native GUI frameworks. We introduce smaps-profiler, a tool to profile the overall memory consumption of Linux applications that can report memory usage other profilers simply do not measure. Using this tool, we conduct experiments which suggest that most of the extra memory usage compared to native applications could be due the size of the web browser program itself. We discuss our experiments and findings, and conclude that even more rigorous studies are needed to profile GUI applications.


Elizabeth Wyss

A New Frontier for Software Security: Diving Deep into npm

When & Where:


Eaton Hall, Room 2001B

Degree Type:

PhD Dissertation 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 end-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 distinctive phenomena that impact its overall security and usability. Such factors include (i) hidden code clones--which may stealthily propagate known vulnerabilities, (ii) install-time attacks enabled by unmediated installation scripts, (iii) hard-coded URLs residing in package code, (iv) the impacts of open-source development practices, (v) package compromise via malicious updates, (vi) spammers disseminating phishing links within package metadata, and (vii) abuse of cryptocurrency protocols designed to reward the creators of high-impact packages. 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. 


Alfred Fontes

Optimization and Trade-Space Analysis of Pulsed Radar-Communication Waveforms using Constant Envelope Modulations

When & Where:


Nichols Hall, Room 246 (Executive Conference Room)

Degree Type:

MS Thesis Defense

Committee Members:

Patrick McCormick, Chair
Shannon Blunt
Jonathan Owen


Abstract

Dual function radar communications (DFRC) is a method of co-designing a single radio frequency system to perform simultaneous radar and communications service. DFRC is ultimately a compromise between radar sensing performance and communications data throughput due to the conflicting requirements between the sensing and information-bearing signals.

A novel waveform-based DFRC approach is phase attached radar communications (PARC), where a communications signal is embedded onto a radar pulse via the phase modulation between the two signals. The PARC framework is used here in a new waveform design technique that designs the radar component of a PARC signal to match the PARC DFRC waveform expected power spectral density (PSD) to a desired spectral template. This provides better control over the PARC signal spectrum, which mitigates the issue of PARC radar performance degradation from spectral growth due to the communications signal. 

The characteristics of optimized PARC waveforms are then analyzed to establish a trade-space between radar and communications performance within a PARC DFRC scenario. This is done by sampling the DFRC trade-space continuum with waveforms that contain a varying degree of communications bandwidth, from a pure radar waveform (no embedded communications) to a pure communications waveform (no radar component). Radar performance, which is degraded by range sidelobe modulation (RSM) from the communications signal randomness, is measured from the PARC signal variance across pulses; data throughput is established as the communications performance metric. Comparing the values of these two measures as a function of communications symbol rate explores the trade-offs in performance between radar and communications with optimized PARC waveforms.