Digital Commons Network^™

Feature Fusion Transferability Aware Transformer For Unsupervised Domain Adaptation, Xiaowei Yu, Zhe Huang, Zao Zhang

Computer Science Faculty Research & Creative Works

Unsupervised domain adaptation (UDA) aims to leverage the knowledge learned from labeled source domains to improve performance on the unlabeled target domains. While Convolutional Neural Networks (CNNs) have been dominant in previous UDA methods, recent research has shown promise in applying Vision Transformers (ViTs) to this task. In this study, we propose a novel Feature Fusion Transferability Aware Transformer (FFTAT) to enhance ViT performance in UDA tasks. Our method introduces two key innovations: First, we introduce a patch discriminator to evaluate the transferability of patches, generating a transferability matrix. We integrate this matrix into self-attention, directing the model to focus …

Ca-Vqvae: Cortical Folding Aware Numerical Representation Of White-Matter Structure, Yanjun Lyu, Jing Zhang, Lu Zhang, Tong Chen, Xiaowei Yu, Minheng Chen, Yan Zhuang, Chao Cao, Tianming Liu, Dajiang Zhu

Computer Science Faculty Research & Creative Works

White matter (WM) serves as a fundamental component of the brain providing essential structural support and facilitating the brain cognitive processes. Thus, an accurate and efficient description of the brain's white matter structure is essential for understanding brain function connectivity and development. In this work we used the deep model to combine the information of the WM fiber bundle shape and its related cortical folding patterns together representing the WM fiber bundle from diffusion MRI tractography into a pre-defined low-dimensional space and generate the numerical representation vector. This cortical-aware vector-quantized variational encoder (CA-VQVAE) framework leverages cortical locations and folding patterns …

Classiffication Of Mild Cognitive Impairment Based On Dynamic Functional Connectivity Using Spatio-Temporal Transformer, Jing Zhang, Yanjun Lyu, Xiaowei Yu, Lu Zhang, Chao Cao, Tong Chen, Minheng Chen, Yan Zhuang, Tianming Liu, Dajiang Zhu

Computer Science Faculty Research & Creative Works

Dynamic functional connectivity (dFC) using resting-state functional magnetic resonance imaging (rs-fMRI) is an advanced technique for capturing the dynamic changes of neural activities and can be very useful in the studies of brain diseases such as Alzheimer's disease (AD). Yet, existing studies have not fully leveraged the sequential information embedded within dFC that can potentially provide valuable information when identifying brain conditions. In this paper, we propose a novel framework that jointly learns the embedding of both spatial and temporal information within dFC based on the transformer architecture. Specifically, we first construct dFC networks from rs-fMRI data through a sliding …

Echopulse: Ecg Controlled Echocardiograms Video Generation, Yiwei Li, Sekeun Kim, Zihao Wu, Hanqi Jiang, Yi Pan, Pengfei Jin, Sifan Song, Yucheng Shi, Xiaowei Yu, Tianze Yang, Tianming Liu, Quanzheng Li, Xiang Li

Computer Science Faculty Research & Creative Works

Echocardiography (ECHO) is essential for cardiac assessments, but its video quality and interpretation heavily rely on manual expertise, leading to inconsistent results from clinical and portable devices. ECHO video generation offers a solution by improving automated monitoring through synthetic data and generating high-quality videos from routine health data. However, existing models often face high computational costs, slow inference, and rely on complex conditional prompts that require experts' annotations. To address these challenges, we propose ECHOPulse, an ECG-conditioned ECHO video generation model. ECHOPulse introduces two key advancements: (1) it accelerates ECHO video generation by leveraging VQ-VAE tokenization and masked visual token …

Secure Data Relay In Federated Digital Twins Of Iot-Enabled Smart Interconnected Factories, Anusha Vangala, Jack Wyeth, Ashok Kumar Das, Sajal K. Das

Computer Science Faculty Research & Creative Works

Smart interconnected factories allow manufacturing units from physically distanced factory sites to communicate classified information needed for additive manufacturing. Each factory has interconnected digital twins of their equipment autonomously managed by a Point-of-Contact digital twin creating a hierarchical system with federated digital twins. The data sharing between the factories must be directed through an edge node responsible for managing multiple factories. We proposed a novel lightweight protocol to prevent the leakage of classified information at any nodes other than the origin and destination digital twins. It leverages elliptic curve cryptography to design a proxy re-encryption scheme with the edge node …

Rush: Rule-Based Scheduling For Low-Latency Serverless Computing, Priyanka Ashok Birajdar, Kush Anchalia, Anurag Satpathy, Sourav Kanti Addya

Computer Science Faculty Research & Creative Works

Serverless computing abstracts server management, enabling developers to focus on application logic while benefiting from automatic scaling and pay-per-use pricing. However, dynamic workloads pose challenges in resource allocation and response time optimization. Response time is a critical performance metric in serverless environments, especially for latency-sensitive applications, where inefficient scheduling can degrade user experience and system efficiency. This paper proposes RUSH (Rule-based Scheduling for Low-Latency Serverless Computing), a lightweight and adaptive scheduling framework designed to reduce cold starts and execution delays. RUSH employs a set of predefined rules that consider system state, resource availability, and timeout thresholds to make proactive, latency-Aware …

Reindsplit: Reinforced Dynamic Split Learning For Pest Recognition In Precision Agriculture, Vishesh Kumar Tanwar, Soumik Sarkar, Asheesh K. Singh, Sajal K. Das

Computer Science Faculty Research & Creative Works

To empower precision agriculture through distributed machine learning (DML), split learning (SL) has emerged as a promising paradigm, partitioning deep neural networks (DNNs) between edge devices and servers to reduce computational burdens and preserve data privacy. However, conventional SL frameworks' one-split-fits-all strategy is a critical limitation in agricultural ecosystems where edge insect monitoring devices exhibit vast heterogeneity in computational power, energy constraints, and connectivity. This leads to straggler bottlenecks, inefficient resource utilization, and compromised model performance. Bridging this gap, we introduce ReinDSplit, a novel reinforcement learning (RL)-driven framework that dynamically tailors DNN split points for each device, optimizing efficiency without …

Fuzzy-Based Deep Reinforcement Learning For Suicidal Ideation Detection In Online Social Networks, Greeshma Lingam, Sajal K. Das

Computer Science Faculty Research & Creative Works

Suicidal ideation is a major psychological problem, and preventing this social risk is recognized as an important research topic. In reality, there can be several reasons why a person experiences suicidal ideation. Each individual can express views, emotions, and several types of symptoms related to suicidal ideation on the most popular social media platforms. In online social networks (OSNs), identification of suicidal ideation is one of the major challenging tasks. Existing studies have shown that the delay in understanding and identifying various risk factors can cause the suicidal event to occur. Due to the scarcity of data and understanding, the …

When Federated Learning Meets Quantum Computing: Survey And Research Opportunities, Aakar Mathur, Ashish Gupta, Sajal K. Das

Computer Science Faculty Research & Creative Works

Quantum Federated Learning (QFL) is an emerging field that harnesses advances in Quantum Computing (QC) to improve the scalability and efficiency of decentralized Federated Learning (FL) models. This paper provides a systematic and comprehensive survey of the emerging problems and solutions when FL meets QC, from research protocol to a novel taxonomy, particularly focusing on both quantum and federated limitations, such as their architectures, Noisy Intermediate Scale Quantum (NISQ) devices, and privacy preservation, so on. With the introduction of two novel metrics, qubit utilization efficiency and quantum model training strategy, we present a thorough analysis of the current status of …

Circa: A Framework For Collaborative Identification Of Root Cause Analysis In Iot Microservices, Xingguo Jiang, Hong Luo, Yan Sun, Sajal K. Das

Computer Science Faculty Research & Creative Works

With continuous growth of IoT applications, service failures are quite inevitable. Due to the complexity and dynamics of IoT services, the root cause analysis (RCA) following an alert can assist in quickly resolving the possible faults. However, the time scales of metrics (e.g., CPU utilization, memory usage) generated by microservices and the dynamic topologies generated by calls between the Application Program Interfaces (APIs) are different. Moreover, the status of devices is an important aspect of RCA in IoT. All these make it extremely challenging to learn failure features of microservice metrics and API calls. Therefore, we propose a novel framework …

Mgco: Mobility-Aware Generative Computation Offloading In Edge-Cloud Systems., Aswini Ghosh, Nelson Sharma, Shivendu Mishra, Rajiv Misra, Sajal K. Das

Computer Science Faculty Research & Creative Works

Mobility introduces significant challenges for optimal computation offloading, latency minimization, and efficient re source utilization in multi-access edge computing (MEC) systems. A key difficulty lies in leveraging real user trajectories to jointly optimize horizontal (inter-edge) and vertical (edge-to-cloud) task offloading decisions. This paper proposes a two-dimensional offloading scheme for a multi-layer edge–cloud architecture that enables collaborative task execution among resource-constrained edge nodes under mobility conditions. We present MGCO (Mobility-Aware Generative Computation Offloading), a generative AI–driven Transformer-based sequence-to-sequence Deep Q-Network (s2s-DQN) framework that learns from real-time trajectory data to anticipate user movement and optimize task placement dynamically. The Transformer architecture is …

Content Subversion Against 1 Information-Based Systems, Junjie Xiong, Ian Markwood, Dakun Shen, Yao Liu, Zhuo Lu

Computer Science Faculty Research & Creative Works

We present a novel class of content subversion attacks against information-based services, causing documents to appear to humans dissimilar to the underlying content extracted by information-based services. We demonstrate the significant impact of these attacks on real-world systems through five distinct variants. Our first attack allows academic paper writers and reviewers to collude via subverting the automatic reviewer assignment systems in current use by academic conferences including INFOCOM, which we reproduced. Our second attack renders ineffective plagiarism detection software, particularly Turnitin, targeting specific small plagiarism similarity scores to appear natural and evade detection. In our third attack, we place masked …

Message From Workshop Chairs, Sushil K. Prasad, Srishti Srivastava, Satish Puri, David Bunde, Shubbhi Taneja, Buddhi Ashan Mallika Kankanamalage

Computer Science Faculty Research & Creative Works

No abstract provided.

Real-Time Testbed For Studying Cyberattacks And Defense In Der-Integrated Smart Inverter Systems, M. Maliha, A. Oluyomi, M. Booge, S. Bhattacharjee, N. Braasch, P. Gomez, Sajal K. Das

Computer Science Faculty Research & Creative Works

In this paper, we propose a Hardware-in-the-Loop (HIL) simulation testbed suitable for the implementation and testing of realistic cyberattacks on grid-tied smart inverter systems integrated with Distributed Energy Resources (DER) that use the Distributed Network Protocol-3 (DNP3) protocol for communications between grid components. Specifically, our testbed combines a Real-Time Digital Simulator (RTDS) NovaCor device, outfitted with GNETx2 network interface cards, a grid-tied DER topology implemented via the RTDS software package RSCAD, and a custom virtual network that emulates a man-in-the-middle (MITM) attacker. The MITM attacker captures DNP3 traffic and falsifies telemetry data in DNP3 packets to trigger unwarranted commands from …

Grace-Fl: Green Resource-Aware Communication-Efficient Federated Learning, Dipanwita Thakur, Antonella Guzzo, Giancarlo Fortino, Sajal K. Das

Computer Science Faculty Research & Creative Works

Federated Learning (FL) enables collaborative model training across distributed clients while preserving data privacy, but its deployment on resource-constrained devices is hindered by high communication overhead, inefficient energy usage, and poor convergence under non-IID data distributions. To address these challenges, we propose GRACE-FL: a Green Resource-Aware Communication-Efficient Federated Learning framework that explicitly incorporates device energy capacity into training. Each client adapts its learning rate, number of local epochs, and gradient quantization bit-width based on its available energy, allowing high-capacity devices to sustain more intensive training while low-capacity devices operate with lighter configurations. A novel energy-weighted aggregation strategy ensures that clients …

Alertble: Alert Workzone Hazards Using Hybrid Filtering And Machine-Learning-Enabled Ble, Samuel Akinyede, Sejun Song

Computer Science Faculty Research & Creative Works

Collision hazard detection in industrial work zones faces challenges from signal instability, mobility-induced fluctuations, and nonline-of-sight (NLOS) conditions. While Bluetooth low energy (BLE) offers cost-effective proximity sensing, its received signal strength indicator (RSSI) variability - fluctuating by ±10 dBm even at fixed distances - limits reliability in safety-critical applications. This article presents AlertBLE, a hybrid BLE-based hazard detection system that combines extended Kalman filter (EKF) and adaptive moving average (AMA) algorithms to achieve up to 94% RSSI variance reduction in static NLOS conditions. The system introduces speed-aware safety thresholds based on reaction time and braking distance models, dynamically expanding hazard …

Should Hate Speech Be Criminalized? Lessons From The Canadian Experience In R V. Zundel And R V. Keegstra, Kenneth Grad

FIU Law Review

There is a global trend toward increased use of criminal law to combat hate speech. In assessing this trend, one should be mindful of the experience of countries that have long had criminal laws targeting harmful expression. Canada is one such country. Using the leading Canadian cases of R v. Zundel and R v. Keegstra, this article argues that the Canadian experience suggests the criminal law is a flawed mechanism for countering harmful expression. This is so for at least three reasons. First, hate-speech prosecutions may undermine the group dignity and sense of inclusion of minority groups. Second, criminal laws …

Confronting The Challenges Of Regulating Artificial Intelligence, Amy B. Cyphert

FIU Law Review

Public opinion polls conclude that the American public is in favor of regulating artificial intelligence (“AI”), and many technology companies publicly claim that they would welcome regulation. And yet the United States has struggled to enact federal comprehensive AI regulations beyond a short-lived Executive Order. Why? Part I of this Article explains why regulating AI is so difficult, focusing on six key reasons: AI is a global issue; AI is not one discrete issue; AI is developing at a speed that is unprecedented; lawmakers largely lack the technical expertise effective AI regulation requires; the stakes of getting the regulation wrong …

Nevada’S Blockchain Gamble: Can A State Embracing Web3 Technology Lead Probate Courts Into The Digital Age?, Ariel Sweeney

FIU Law Review

Probate stands as a bastion of legal formalism, seemingly resistant to the transformative currents of digital innovation that have swept through other domains of American law. While financial transactions, real property conveyances, and contract execution have increasingly begun exploring the use of Web3 technologies such as blockchain and smart contracts, estate and probate law remain tethered to paper-based procedures and rigid execution requirements. Nevada was the first state to provide legal support for Web3 technology, amending its Uniform Electronic Transactions Act statutes in 2017 to recognize blockchain-based transactions as valid and judicially enforceable. Yet despite this progressive legislative framework, the …

Electrochemically Active Molybdenum Phosphate With Open-Framework Structure, Sutapa Bhattacharya, Milad Aghayi-Anaraki, Nikolay Gerasimchuk, Steven P. Kelley, Amitava Choudhury

Chemistry Faculty Research & Creative Works

Somewhat less investigated, lithium ion containing molybdenum phosphate compounds are synthesized via simple template free hydrothermal method. Single-crystal X-ray diffraction (SCXRD) study reveals that the compounds Li_0.89Mo₂O_1.89F_0.11(PO₄)2(H₂PO₄)·xH₂O (x ≈1) and Li₃Mo₂O₂(PO₄)₃·xH₂O (x ≈1) crystallize in monoclinic C2/m and C2/c space group, respectively. The structures are built of alternate corner-sharing of MoO₆ octahedra and PO₄ tetrahedra forming layers, which are pillared by phosphate groups creating a 3D framework with channels. These …

Atomistic Mechanisms Of Ti3alc2 Etching: Oxidation, Surface Stability, And Selectivity, Valentina Nesterova, Ana Maria Stratulat, Walter Malone, Vadym Mochalin, Majid Beidaghi, Konstantin Klyukin

Chemistry Faculty Research & Creative Works

Mechanisms governing MAX-phase etching set the properties and stability of resulting MXenes, yet the details of this complex process remain poorly understood. Combining thermodynamic analysis of surface chemistries with ab initio molecular dynamics and enhanced free-energy sampling, we directly model elementary steps at the MAX/etching solution interface. We find that purely thermodynamic analysis overestimates etching selectivity, which is governed by the kinetics of elementary reaction steps. Our calculations reveal that initial exposure to water drives localized, thermodynamically favored edge oxidation that seeds MXene-like terminations into the uppermost layers of Ti₃AlC₂ with oxygen occupying the interlayer space between …

Impact Of Surface And Physical Property On Multiphase Flow In Sealed Vessel: Liquid Dropdown Performance, Mehedi Hasan Tusar, Palash K. Bhowmik, Kazuma Kobayashi, Syed Bahauddin Alam, Shoaib Usman

Nuclear Engineering and Radiation Science Faculty Research & Creative Works

This study explores multi-phase (i.e., liquid-gas) and multi-fluid (i.e., air-water, and water-silicone oil) flow-pattern and flow-blockage physics phenomena for wall wettability conditions ranging from superhydrophilic to superhydrophobic cases in sealed vessels utilizing computational fluid dynamics (CFD) simulation tools and volume-of-fluid (VOF) method with sharp interface modeling. Detailed modeling and simulation (M&S) of such physics phenomena—in which liquid (e.g., water) stands over top of gas (e.g., air or steam) in a closed channel and exhibited flow blockage, flow reversal related challenges—are pivotal for design, analysis, and qualification of component-level (e.g., heat pipes, heat exchangers) to system-level (e.g., emergency core cooling systems …

Joint 3d Beamforming-And-Trajectory Design For Uav-Satellite Uplink Covert Communication, Jihong Yu, Yuting Cai, Shihao Yan, Yun Li, Jingjing Wang, Jiahao Liu, Jianping An

Research outputs 2022 to 2026

In this paper,we study uplink covert communication in a space-air system,where an unmanned aerial vehicle (UAV) transmits sensitive data to a Geosynchronous Earth Orbit (GEO) satellite while preventing the transmission action from being discovered by a warden. We derive the optimal decision threshold of the warden. We investigate the 3-dimensional (3D) beamformer and 3D trajectory design for the transmitter UAV against this optimum warden to maximize the covert transmission rate in the presence of imperfect channel state information and uncertain noise. Due to the non-convex structure and dependence between beamforming vectors and locations of the transmitter UAV,we develop a decoupling …

System Dynamics With Insight Maker, Steven D'Alessandro, Fons Wijnhoven

Research outputs 2022 to 2026

This book offers a practical, model-driven pathway for reasoning about uncertain futures in business and public policy using system dynamics with Insight Maker. It begins by motivating why historical data alone often fail to predict social change, and it introduces the core language of system dynamics—stocks, flows, feedbacks, delays, and auxiliary variables—alongside the complementary use of agent-based modeling. Through business-relevant cases (e.g., park management trade-offs, epidemic–economy interactions, and industry competition), the book demonstrates how non-linear structure generates counter-intuitive dynamics, why scenario analysis is essential, and how to translate causal loop diagrams into stock-and-flow simulations. Readers are guided step-by-step to build, …

Developing Memory And Understanding Skills For Subject-Specific Knowledge Processing And To Face High-Stakes Exams: A Case Study, Jabbar Al Muzzamil Fareen

Journal of Educational and Psychological Studies

University students possess different levels of language and cognitive abilities in gaining their subject knowledge and face tremendous challenges to get through the exams in distinction. The present education system expects the students to possess excellent subject knowledge and outstanding results in their exams. In light of this context, this paper attempts to address how students are engrossed in learning difficulties while subject knowledge processing and facing their examinations. Qualitative case study approach has been undertaken to analyze the language and cognitive skills of the students in memory processing of their subject knowledge to face the exams. Their language learning …

Corporate Scenarios: Drawing Lessons From History, Madison Condon

Faculty Scholarship

As corporations are increasingly pressed to reveal information about their exposure to climate-related risks, they are often asked to undertake and disclose the outcome of “scenario analysis.” In this exercise, corporations, including financial institutions, examine how their business would fare under different pathways the future may take. One oft-used scenario, for example, is the International Energy Agency’s “Net-Zero by 2050: A Roadmap for the Energy Sector.” This Essay presents a history of the use of scenarios as a corporate planning tool, particularly in the oil industry, arguing that it is key for understanding our present moment and the role of …

The Complex And Marginalized Experiences Of Bipoc Trafficked Women: An Examination Of Disabilities, Aces, Discrimination, And Racism, Jacquelyn C. A. Meshelemiah, Fabian Arroyo Rojas, Hannah Ruth Steinke, Marlene Carson, Justin A. Haegele

Human Movement Studies & Special Education Faculty Publications

The purpose of this study was to examine the lived experiences and multiple identities of disabled BIPOC trafficked women. The findings from this study help to identify a carousel of victimization experienced by disabled BIPOC trafficked women, starting with adverse childhood experiences, onto trafficking victimization that differed between Black and White women, and later while seeking services. These findings highlight the need for providers and researchers to think beyond monolithic identities and consider the intersecting ways in which various forms of oppression (ableism and racism) influence the experiences of disabled trafficked BIPOC women.

Toward Embodied Navigation Through Vision And Language, Muraleekrishna Gopinathan

Theses: Doctorates and Masters

Embodied AI is a challenging but exciting field in which a robot learns to interact with human-living spaces to perform various tasks. This thesis studies the embodied navigation problem in which a robotic agent navigates in a previously unseen indoor environment based on a challenging task. In particular, the Vision-and-Language Navigation (VLN) task requires a robot to navigate based on a descriptive human-language instruction. This thesis aims to improve VLN agents on four key aspects - their understanding of the environment, training via additional data, correcting navigational errors, and predicting the layout of the environment for better planning.

First, we …

The Rise Of Women In Mining - How To Improve Inclusion And Diversity Through Organisational Programs, Ashley Mcgrath

Theses: Doctorates and Masters

The inclusion and diversity (I&D) of women has become a priority for organisations due to the positive association with business performance and rising expectations of employees and shareholders for fair outcomes for all genders. Improving the I&D of women is also being driven by the evolving Australian legal framework which has heightened the responsibility of organisations regarding psychological safety and the prevention of sexual harassment and sex-based discrimination. Companies operating in male dominated industries in Western Australia (WA) are under particular pressure due to lagging performance in addressing equity issues at a state level. Existing research correlates the I&D of …

Embodied Ai For Challenging Rearrangement Tasks In The Context Of Service And Assistive Robots, Mariia Khan

Theses: Doctorates and Masters

Embodied AI explores intelligent agents that learn through interaction with their environment, aiming to replicate human-like learning processes. Achieving this requires agents capable of understanding a scene via various sensors, reasoning about their actions, and reacting accordingly. These abilities are necessary for service domestic robots to assist humans in their day-to-day activities. Embodied AI tasks can include but are not limited to: visual exploration, visual navigation, instruction following and embodied question answering, which typically consider static (unchanging) environments, where objects do not move over time. This thesis addresses one of the most challenging Embodied AI tasks – visual room rearrangement, …