Author: Guilbeault, Jessica

Center for Voting Technology Conducts Comprehensive Assessment of New Connecticut Voting Machines

The Secretary of State asked UConn's VoTeR Center to test multiple voting tabulators for reliability and evaluate which models are least prone to cyberattacks

Members of UConn's VoTeR Center team work to ensure election outcomes conducted with electronic voting systems are secure and dependable.

This election year, UConn’s College of Engineering (CoE) is helping to ensure trust in every vote cast.

Last month, Connecticut Secretary of the State Stephanie Thomas announced 2,700 paper-based voting tabulators, statewide, will be replaced with new, state-of-the-art machines. The state hasn’t upgraded most voting equipment in almost 18 years.

“This is a pivotal moment for Connecticut’s elections, and one that is a long time in the making,” Thomas said in a recent press release. “Through this milestone tabulator upgrade, we’re providing our election administrators with the modern tools they need to run efficient elections.”

Since choosing the safest, most reliable tabulators was a crucial step in the replacement process, Thomas turned to the CoE’s Voting Technology Research (VoTeR) Center for guidance. Since 2006, members of the VoTeR Center have strived to assess the security and dependability of electronic voting equipment and develop new techniques for auditing the results of elections.

Laurent Michel, technical director of the VoTeR Center and professor of computer science and engineering.
Laurent Michel is technical director of the VoTeR Center and professor of computer science and engineering (UConn Photo).

“For this evaluation, the VoTeR Center devised and executed testing procedures meant to assess the resilience of potential tabulators and the eco-system in which they operate against adversarial attacks,” explains Laurent Michel, technical director of the VoTeR Center and professor of computer science and engineering. “White-hat ethical hacking of this type is meant to find weaknesses in the equipment, or the processes election officials rely on to program, execute, and tabulate results state-wide.”

Over several weeks, the VoTeR team worked to evaluate potential new tabulators on the basis of cybersecurity guarantees, support for best-practice election audits, and compliance with the Voluntary Voting System Guidelines set by the U.S. Election Assistance Commission. All findings inform officials as to the ideal safe-used processes that should be adopted to conduct elections with secure tabulators, Michel says.

Ultimately, the VoTeR team shared their evaluations with Thomas and the selection committee, and the State began purchasing the equipment. Secretary Thomas plans to distribute the new machines to nine Connecticut towns prior to the November general election. Other towns will receive theirs in 2025.

“Such an evaluation touches on many technical issues ranging from compliance to the standards to resilience to attacks an adversary might be tempted to carry out against a voting system, such as tampering with the equipment to coerce it into reporting incorrect results,” Michel says.

Michel, a founding member of the VoTeR Center, also serves as director of UConn’s Synchrony Financial Center of Excellence in Cybersecurity and co-director of the Connecticut Cybersecurity Center. At VoTeR, he works alongside Center Director Alexander Russell, professor of computer science and mathematics, Benjamin Fuller, associate professor of computer science, and several research software engineers, faculty, graduate, and undergraduate assistants. All three faculty teach in the CoE’s School of Computing.

“While directly supporting the State, the Center also pursues research in election integrity and auditing, with active involvement of undergraduates and graduate students,” Russell says.

The VoTer Center was formed in response to the Help America Vote Act, signed into law in 2002, and initially helped the State select the very tabulators that are currently at end of life. Since then, the purview of the center has significantly expanded, now supporting the State’s annual hand-counted audit procedures, providing forensic audits of electronic tabulators, developing technological tools for ballot processing and verification of voter assignments, and playing a critical role in the State’s efforts to guarantee voting rights.

“Proper auditing not only increases the confidence of the voters that state elections are run, but it also helps uncover procedural failings of the election process, enabling voting districts to better serve their constituents,” Michel says. “Our goals are to ensure the integrity of the election outcomes conducted with electronic voting systems and to continuously assess their security and dependability.”

View other reports, publications, and methodologies the Center relies on here.

Professors Creating Computing Models to Increase Public Trust During Elections

The team will report on optimizing the standard for auditing election reporting, analyzing elections and their components, and specifying procedures for desired security.

School of Computing professor Benjamin Fuller discussing the Secure, Holistic Infrastructure for Election Logistics and Data (SHIELD) project at the University of Nebraska at Omaha in October.

UConn Engineering professors are aiding a national effort to maintain secure election infrastructure, ensuring fair elections for all United States voters.

School of Computing professors Benjamin Fuller, Laurent Michel, Ghada Almashaqbeh, and Alexander Russell partnered with the University of Nebraska at Omaha to launch the Secure, Holistic Infrastructure for Election Logistics and Data (SHIELD) project in October. The SHIELD project is supported by the National Counterterrorism Innovation, Technology, and Education (NCITE) Center, a United States Department of Homeland Security Center of Excellence.

The election system in the United States is historically complex, with local and state offices given decentralization and autonomy. The system offers oversight and independence to local and state offices. This decentralization yields increased cybersecurity resilience. However, the lack of sharing can result in duplication of efforts or a waste of limited resources.

This project will develop tools and processes that solidify the decentralized electoral systems in the United States to increase the trust of stakeholders in election outcomes. The research team will design a non-prescriptive formal process for election officials to reason holistically about the security of elections.

SHIELD has two main goals, including reporting on optimizing the standard for auditing election reporting, analyzing elections and their components, and specifying procedures for desired security; and organizing an Omaha forum on election security.

Fuller visited the University of Nebraska at Omaha early in October to attend an event hosted by NCITE, which brought Jen Easterly, director of the Cybersecurity and Infrastructure Security Agency, and five Midwestern secretaries of state to discuss the challenges of the 2024 election and priorities for keeping it secure.

“By partnering with the University of Nebraska at Omaha, we can elevate our impact and continue to provide thoughtful models for election audits and secure systems,” Fuller says. “Boosting the public’s trust in the electoral process should be considered a key offering from a public institution like UConn.”

The four UConn researchers have experience in applied cryptography, cryptography, computer systems security, privacy, information theory, modeling and programming languages, combinatorial optimization, constraint programming, electronic voting security, and statistical election auditing.

This project is one of many UConn is leading related to election standards and national security.

“Our faculty in the School of Computing are recognized authorities in their respective domains, and they are profoundly dedicated to strengthening the integrity of electoral processes,” says School of Computing Director Sanguthevar Rajasekaran. “Their pioneering research on voting security and election standards plays a crucial role in fortifying the resilience and reliability of our democratic systems. I take great pride in working with such distinguished scholars committed to advancing this essential field of study.”

Read more about the SHIELD project online.

We are Pleased to Welcome Rahul Jayachandran and Joel Duah For Our Summer 2024 REU Student

“Hi, I’m Rahul Jayachandran and I’m a rising college sophomore from Glastonbury, CT. I am pursuing a degree in computer science and math and am excited to spend this summer conducting research at UConn. In my free time I enjoy rowing and playing piano.”

 

“Hello, I’m Joel Duah, a junior at UConn from Manchester, CT. I’m pursuing a major in Computer Science and Engineering with a concentration in Computational Data Analytics. I’m enthusiastic about applying my knowledge from past semesters to this research opportunity, particularly in exploring voting patterns in Connecticut. Through this REU, I aim to increase my proficiency in handling data and making it more accessible to individuals without technical backgrounds. Outside of academics, I find joy in hiking and reading.”

 

Congratulations!

Congratulations Sheida Nabavi on your SPARK Grant

 

Funding Agency: UConn FY24 SPARK Technology Commercialization Fund 

Title: AI-CAD for Breast Cancer Screening 

Amount: $50,000 (with the possibility of an additional $50,000)

Dates: March 1, 2024 to April 30, 2025

Congratulations Caiwen Ding on your Amazon Grant

 

Funding Agency: Amazon

Title: Graph of Thought: Boosting Logical Reasoning in Large Language Models

Amount: $70,000 Cash + $50,000 AWS Credits

Dates: April 2024- March 2025

CyberSEED 2024

We had another great CyberSEED event this past Saturday March 23, 2024 with 96 teams with 226 students. We had some intense competition with the top team being the only one to solve all of the challenges. The briefing presentations ended up being a pretty deciding factor in the placement of the top teams and is such a valuable component of the experience for the students.

Congratulations to the top 10 teams, UConn coming in 2nd place!
Award Ceremony Presentation

Congratulations Caiwen Ding and Dongjin Song on your NSF CAREER Awards!

Caiwen Ding

Caiwen Ding

Congratulations Caiwen Ding on receiving a National Science Foundation CAREER Award for his proposal titled “CAREER: Algorithm-Hardware Co-design of Efficient Large Graph Machine Learning for Electronic Design Automation”. The goal is the project is to address the efficiency and scalability of using graph learning for Electronic Design Automation, thought a series of algorithm-hardware codesign approaches.

Caiwen Ding is an assistant professor in the School of Computing at the University of Connecticut. He received his Ph.D. degree from Northeastern University (NEU), Boston in 2019,  supervised by Prof. Yanzhi Wang.  His interests include Algorithm-system co-design of machine learning/artificial intelligence, privacy-preserving machine learning, machine learning for electronic design automation (EDA), and neuromorphic computing. He is a recipient of the 2024 CISCO Research Award and NSF CAREER Award. He received the best paper nomination at 2018 DATE and 2021 DATE, the best paper award at the DL-Hardware Co-Design for AI Acceleration (DCAA) workshop at 2023 AAAI, outstanding student paper award at 2023 HPEC, publicity paper at 2022 DAC, and the 2021 Excellence in Teaching Award from UConn Provost. His team won first place in accuracy and fourth place overall at the 2022 TinyML Design Contest at ICCAD. He was ranked among Stanford’s World’s Top 2% Scientists in 2023. His research has been mainly funded by NSF, DOE, DOT, USDA, SRC, and multiple industrial sponsors.

Abstract: Estimating Power, Performance, and Area (PPA) earlier in the electronic design automation (EDA) flow would improve the Quality of Results (QoR) and reliability in chip design. The classical analytical or heuristic methods can be challenging to fine-tune, especially for complex problems. Machine learning (ML) methods have proven to be effective in addressing these problems. Graph Neural Networks (GNNs) have gained popularity since they are among the most natural ways to represent the fundamental objects in the EDA flow. However, with increased design complexity and chip capacity, an increasing performance gap exists between the extremely large graphs in EDA and the insufficient support from general-purpose hardware, such as mainstream graphics processing units (GPUs). This project aims to expedite the large graph machine learning on various EDA tasks, through a full-fledged development of efficient and scalable computing paradigms. This project's novelties are EDA domain knowledge-aware graph machine learning, training acceleration, and algorithm-hardware co-design and optimization. The project's broader significance and importance include: (1) to advance the field of machine learning in chip design, highlighted in National Artificial Intelligence Initiative; (2) to deepen the understanding of interactions among EDA domain knowledge, graph learning, and GPU acceleration; (3) to enrich the computer engineering curriculum and promote participation from undergraduates, underrepresented groups, and K-12 students in STEM fields through relevant programs.

Dongjin Song

Dongjin Song

Project Framework

Dongjin Song

Congratulations Dongjin Song on receiving the prestigious National Science Foundation (NSF) CAREER Award to support his research project titled "CAREER: Towards Continual Learning on Evolving Graphs: from Memorization to Generalization". This project will develop a generic machine learning paradigm, Continual Learning on Evolving Graphs (CoLEG), to resolve the catastrophic forgetting problem by retaining essential structural information and temporal dynamics, ensure the generalization capability, and address real-world applications on evolving graphs. Specifically, he not only plans to tackle the catastrophic forgetting issue in structural evolving graphs via graph sparsification and topology-aware embedding, but also aims to develop new algorithms to incorporate structural and temporal dynamic patterns of evolving graphs under different regimes, resolve the task-free challenge, and reveal high-order dependencies. He will also develop novel solutions to pursue and imp pre-trained models and facilitate test-of-time adaptation to ensure the generalization over unforeseen scenarios.

Dongjin Song has been an assistant professor in the School of Computing, University of Connecticut since Fall 2020. He was previously a research staff member at NEC Labs America in Princeton, NJ. He received his Ph.D. degree in the ECE Department from the University of California San Diego (UCSD) in 2016. His research interests include machine learning, data science, deep learning, and related applications for time series data analysis and graph representation learning. Papers describing his research have been published at top-tier data science and artificial intelligence conferences, such as NeurIPS, ICML, KDD, ICDM, SDM, AAAI, IJCAI, ICLR, CVPR, ICCV, etc. He is an Associate Editor for Neurocomputing and has served as Senior PC for AAAI, IJCAI, and CIKM. He received the prestigious NSF CAREER award in 2024 and the UConn Research Excellence Research (REP) Award in 2021. He has co-organized the AI for Time Series (AI4TS) Workshop at IJCAI, AAAI, ICDM, SDM, and MiLeTS workshops at KDD.

Abstract: In the modern big data era, data often grows continuously and its interconnections and temporal dynamics evolve. To cope with the continuous evolution in data, an intelligent agent needs to incrementally acquire, perceive, accumulate, and exploit structural and temporal dynamic knowledge throughout its lifetime. This project aims to develop a generic machine learning paradigm to conduct Continual Learning on Evolving Graphs (CoLEG). The success of this project will 1) benefit critical infrastructure (such as social networks, transportation, and renewable energy) and human welfare (in the form of, for example, improvements in healthcare and epidemiology), 2) provide an ideal platform for composing the areas of graph representation learning, time series analysis, continual learning, and causal analysis, and 3) develop open-source tools for evolving graphs that can advance diverse topics such as node classification, link prediction, and temporal forecasting, improve our knowledge of the physical world, and contribute to real-world applications. This project will also 1) engage high school students in research and outreach to K-12 teachers and students, 2) broaden the participation of underrepresented groups especially female and low-income students in STEM, and 3) educate undergraduate and graduate students through the development of new course modules in data mining and machine learning.