go-research

Georgia Tech Celebrates 2025 Ph.D. Graduates in Cybersecurity and Privacy

dgivens8 — Mon, 28 Jul 2025 14:47:40 +0000

Georgia Tech Celebrates 2025 Ph.D. Graduates in Cybersecurity and Privacy dgivens8 Mon, 07/28/2025 - 10:47

The School of Cybersecurity and Privacy at Georgia Tech is proud to recognize the accomplishments of five doctoral students who finished their doctoral programs in Spring 2025. These scholars have advanced critical research in software security, cryptography, and privacy, collectively publishing 34 papers, most of which appear in top-tier venues.

Ammar Askar developed new tools for software security in multi-language systems, including a concolic execution engine powered by large language models. He highlighted DEFCON 2021, which he attended with the Systems Software and Security Lab (SSLab), as a favorite memory.

Zhengxian He persevered through the pandemic to lead a major project with an industry partner, achieving strong research outcomes. He will be joining Amazon and fondly remembers watching sunsets from the CODA building.

Stanislav Peceny focused on secure multiparty computation (MPC), designing high-performance cryptographic protocols that improve efficiency by up to 1000x. He’s known for his creativity in both research and life, naming avocado trees after famous mathematicians and enjoying research discussions on the CODA rooftop.

Qinge Xie impressed faculty with her adaptability across multiple domains. Her advisor praised her independence and technical range, noting her ability to pivot seamlessly between complex research challenges.

Yibin Yang contributed to the advancement of zero-knowledge proofs and MPC, building toolchains that are faster and more usable than existing systems. His work earned a Distinguished Paper Award at ACM CCS 2023, and he also served as an RSAC Security Scholar. Yang enjoyed teaching and engaging with younger students, especially through events like Math Kangaroo.

Faculty mentors included Regents’ Entrepreneur Mustaque Ahamad, Professors Taesoo Kim and Vladimir Kolesnikov, and Assistant Professor Frank Li, who played vital roles in guiding the graduates’ research journeys.

Learn more about the graduates and their mentors on the 2025 Ph.D. graduate microsite.

Subtitle

Summary sentence

Summary

Five Ph.D. students from Georgia Tech’s School of Cybersecurity and Privacy graduated in Spring 2025, contributing significant research in software security, cryptography, and privacy with 34 published papers. Highlights include Ammar Askar’s concolic execution tools, Zhengxian He’s industry collaboration, Stanislav Peceny’s advances in multiparty computation, Qinge Xie’s adaptability across domains, and Yibin Yang’s award-winning work on zero-knowledge proofs. Faculty mentors played key roles in supporting their success.

Dateline

Wed, 05/14/2025 - 12:00

jpopham3@gatech.edu

Contact

JP Popham, Communications Officer II

College of Computing | School of Cybersecurity and Privacy

Keywords

students

graduates

PhD

PhD Students

Research

go-research

go-researchnews

Daily Digest

website

academic excellence

Categories

Alumni

Community

Computer Science/Information Technology and Security

Research

Student Honors and Achievements

Student Research

Mercury ID

682394

Source updated

Mon, 07/28/2025 - 10:46

Despite Equal Publication Success in Climate Science, Women Leave the Field Earlier Than Men

admin — Mon, 14 Jul 2025 17:26:46 +0000

Despite Equal Publication Success in Climate Science, Women Leave the Field Earlier Than Men admin Mon, 07/14/2025 - 13:26

A study led by a Georgia Tech researcher analyzing more than a century of climate science publications has found that women in the field are just as productive and successful as their male peers. However, they tend to have shorter careers and therefore fewer total publications.

According to the study, women are 90% as likely as men to maintain their careers.

The research offers one of the deepest looks at publications and gender dynamics in climate science. The field is unique because it blends male-dominated areas of study — geosciences and engineering — with those that are majority-women, including sociology, public health, and life sciences.

The study examined more than 400,000 publications between 1903 and 2018. The findings don’t include topics such as salaries, number of women in the field, or tenure rates.

Read the entire story on the College of Engineering news page.

Subtitle

A century-long analysis of publication data looks at the gender dynamics in a field that combines male- and female-dominated focus areas.

Summary sentence

A study has found that women in climate science are just as productive and successful as their male peers when it comes to journal publications, but are 90% as likely as men to maintain their careers.

Summary

Dateline

Mon, 07/14/2025 - 12:00

maderer@gatech.edu

Contact

Jason Maderer
College of Engineering Communications
maderer@gatech.edu

Associated importer

Keywords

go-research

go-researchnews

News room topics

Earth and Environment

Mercury ID

683122

Source updated

Mon, 07/14/2025 - 13:25

National Report Urges FAA to Overhaul Air Traffic Controller Hiring and Training

admin — Tue, 01 Jul 2025 13:46:41 +0000

National Report Urges FAA to Overhaul Air Traffic Controller Hiring and Training admin Tue, 07/01/2025 - 09:46

A new congressionally mandated report coauthored by a Georgia Tech professor suggests that the Federal Aviation Administration (FAA) hires more air traffic controllers each year, optimizes scheduling, and offers other steps to address a nationwide shortage of the critical staffers.

Karen Feigh and the 13-person committee found that the FAA hired only about two-thirds of the controllers it projected from 2013 to 2023. Due to attrition during that time — and because hiring didn’t accelerate until 2024 — 19 of the FAA’s largest facilities have 15% fewer people managing airspace than they need.

For example, the report shows the tower at Hartsfield-Jackson Atlanta International Airport, the world’s busiest, is 17% below full staff.

The National Academies of Sciences, Engineering, and Medicine report provides guidance to the FAA about establishing appropriate staffing levels. It also suggests specific improvements in hiring, training, scheduling, and fatigue management for the FAA’s 313 facilities.

Read a Q&A with Feigh about the report on the College of Engineering website.

Subtitle

Aerospace engineering professor Karen Feigh cowrote a National Academies review of a national shortage that impacts 45,000 daily commercial flights.

Summary sentence

A new report offers other steps to the Federal Aviation Administration to address a nationwide shortage of the critical staffers.

Summary

Dateline

Mon, 06/30/2025 - 12:00

maderer@gatech.edu

Contact

Jason Maderer
College of Engineering Communications

Associated importer

Keywords

go-researchnews

go-research

News room topics

Science and Technology

Categories

Aerospace

Mercury ID

682947

Source updated

Tue, 07/01/2025 - 09:46

Turning to CubeSats in the Search for Life Thousands of Light-Years from Earth

admin — Tue, 20 May 2025 16:30:33 +0000

Turning to CubeSats in the Search for Life Thousands of Light-Years from Earth admin Tue, 05/20/2025 - 12:30

A new NASA-funded project will have Georgia Tech aerospace engineers developing new technology to one day study planets outside our solar system.

It's a $10 million joint mission led by the University of Michigan called STARI — STarlight Acquisition and Reflection toward Interferometry. Georgia Tech’s engineers will build the propulsion systems for a pair of briefcase-sized CubeSats that will fly in orbit a few hundred yards away from one another, bouncing starlight back and forth.

The technology could be used someday to better understand if any known exoplanets are capable of supporting life as we know it.

Interferometry is already used to study stars, gas clouds, and galaxies. Instead of using one large telescope, several smaller telescopes work as a team. The machines swap starlight to create higher resolution images than are possible from a single telescope.

Scientists and engineers have recently proposed using interferometry to locate exoplanets.

STARI will determine if the same type of coordination and light transmission can be done using less expensive CubeSats.

Read the entire story on the College of Engineering website.

Subtitle

Georgia Tech plays a starring role in NASA’s STARI mission to determine if telescope technology that studies exoplanets can be implemented in briefcase-sized spacecraft.

Summary sentence

A new NASA-funded project will have Georgia Tech aerospace engineers developing new technology to one day study planets outside our solar system.

Summary

A new NASA-funded project will have Georgia Tech aerospace engineers developing new technology to one day study planets outside our solar system.

Dateline

Thu, 02/13/2025 - 12:00

Contact

Jason Maderer
College of Engineering
maderer@gatech.edu

Associated importer

Keywords

go-research

go-researchnews

Categories

Aerospace

Mercury ID

680480

Source updated

Thu, 02/13/2025 - 15:42

SKYSCENES Dataset Could Lead to Safe, Reliable Autonomous Flying Vehicles

admin — Mon, 19 May 2025 13:31:53 +0000

SKYSCENES Dataset Could Lead to Safe, Reliable Autonomous Flying Vehicles admin Mon, 05/19/2025 - 09:31

Is it a building or a street? How tall is the building? Are there powerlines nearby?

These are details autonomous flying vehicles would need to know to function safely. However, few aerial image datasets exist that can adequately train the computer vision algorithms that would pilot these vehicles.

That’s why Georgia Tech researchers created a new benchmark dataset of computer-generated aerial images.

Judy Hoffman, an assistant professor in Georgia Tech’s School of Interactive Computing, worked with students in her lab to create SKYSCENES. The dataset contains over 33,000 aerial images of cities curated from a computer simulation program.

Hoffman said sufficient training datasets could unlock the potential of autonomous flying vehicles. Constructing those datasets is a challenge the computer vision research community has been working for years to overcome.

“You can’t crowdsource it the same way you would standard internet images,” Hoffman said. “Trying to collect it manually would be very slow and expensive — akin to what the self-driving industry is doing driving around vehicles, but now you’re talking about drones flying around.

“We must fix those problems to have models that work reliably and safely for flying vehicles.”

Many existing datasets aren’t annotated well enough for algorithms to distinguish objects in the image. For example, the algorithms may not recognize the surface of a building from the surface of a street.

Working with Hoffman, Ph.D. student Sahil Khose tried a new approach — constructing a synthetic image data set from a ground-view, open-source simulator known as CARLA.

CARLA was originally designed to provide ground-view simulation for self-driving vehicles. It creates an open-world virtual reality that allows users to drive around in computer-generated cities.

Khose and his collaborators adjusted CARLA’s interface to support aerial views that mimic views one might get from unmanned aerial vehicles (UAVs).

What's the Forecast?

The team also created new virtual scenarios to mimic the real world by accounting for changes in weather, times of day, various altitudes, and population per city. The algorithms will struggle to recognize the objects in the frame consistently unless those details are incorporated into the training data.

“CARLA’s flexibility offers a wide range of environmental configurations, and we take several important considerations into account while curating SKYSCENES images from CARLA,” Khose said. “Those include strategies for obtaining diverse synthetic data, embedding real-world irregularities, avoiding correlated images, addressing skewed class representations, and reproducing precise viewpoints.”

SKYSCENES is not the largest dataset of aerial images to be released, but a paper co-authored by Khose shows that it performs better than existing models.

Khose said models trained on this dataset exhibit strong generalization to real-world scenarios, and integration with real-world data enhances their performance. The dataset also controls variability, which is essential to perform various tasks.

“This dataset drives advancements in multi-view learning, domain adaptation, and multimodal approaches, with major implications for applications like urban planning, disaster response, and autonomous drone navigation,” Khose said. “We hope to bridge the gap for synthetic-to-real adaptation and generalization for aerial images.”

Seeing the Whole Picture

For algorithms, generalization is the ability to perform tasks based on new data that expands beyond the specific examples on which they were trained.

“If you have 200 images, and you train a model on those images, they’ll do well at recognizing what you want them to recognize in that closed-world initial setting,” Hoffman said. “But if we were to take aerial vehicles and fly them around cities at various times of the day or in other weather conditions, they would start to fail.”

That’s why Khose designed algorithms to enhance the quality of the curated images.

“These images are captured from 100 meters above ground, which means the objects appear small and are challenging to recognize,” he said. “We focused on developing algorithms specifically designed to address this.”

Those algorithms elevate the ability of ML models to recognize small objects, improving their performance in navigating new environments.

“Our annotations help the models capture a more comprehensive understanding of the entire scene — where the roads are, where the buildings are, and know they are buildings and not just an obstacle in the way,” Hoffman said. “It gives a richer set of information when planning a flight.

“To work safely, many autonomous flight plans might require a map given to them beforehand. If you have successful vision systems that understand exactly what the obstacles in the real world are, you could navigate in previously unseen environments.”

For more information about Georgia Tech Research at ECCV 2024, click here.

Summary sentence

Georgia Tech researchers created a new benchmark dataset of computer-generated aerial images that could allow autonomous flying vehicles to operate reliably and safely.

Summary

Few aerial image datasets exist that can adequately train the computer vision algorithms that would pilot autonomous flying vehicles. Judy Hoffman, an assistant professor in Georgia Tech’s School of Interactive Computing, worked with students in her lab to create SKYSCENES. The dataset contains over 33,000 aerial images of cities curated from a computer simulation program.

Dateline

Thu, 09/26/2024 - 12:00

Contact

Nathan Deen

Communications Officer

School of Interactive Computing

Associated importer

Keywords

go-ai

go-research

Artifical Intelligence

Center for Machine Learning

autonomous drones

drones; UAV; unmanned aerial vehicles

autonomous aircraft

computer vision

computer vision algorithm

computer vision systems

Categories

City Planning, Transportation, and Urban Growth

Computer Science/Information Technology and Security

Research

Mercury ID

677158

Source updated

Thu, 09/26/2024 - 15:12

Rule the Pool This Summer and Make the Biggest Splash

admin — Fri, 16 May 2025 16:37:41 +0000

Rule the Pool This Summer and Make the Biggest Splash admin Fri, 05/16/2025 - 12:37

Want to create the biggest splash in the pool this summer? Forget the bellyflop and the cannonball.

“Popping the Manu” will make you a winner.

Georgia Tech researchers studied dives by the Māori, the indigenous people of New Zealand, who have made Manu jumping a cultural tradition. By hitting the water in a “V” shape, then quickly extending their bodies underwater, they’ve perfected the art of huge splashes.

See a video on how to make the splash and read the entire story on the College of Engineering homepage.

Subtitle

Georgia Tech roboticists explain the physics of epic pool jumps and the New Zealanders who have mastered them

Summary sentence

By hitting the water in a “V” shape, then quickly extending their bodies underwater, the Māori have perfected the art of huge splashes.

Summary

Dateline

Fri, 05/16/2025 - 12:00

maderer@gatech.edu

Contact

Jason Maderer
College of Engineering
maderer@gatech.edu

Associated importer

Keywords

go-researchnews

go-research

News room topics

Science and Technology

Categories

Research

Robotics

Mercury ID

682424

Source updated

Fri, 05/16/2025 - 12:37

Kinaxis and AI4OPT at Georgia Tech Announce Co-Innovation Partnership to Advance Scalable AI in Supply Chain Orchestration

admin — Wed, 02 Apr 2025 15:27:41 +0000

Kinaxis and AI4OPT at Georgia Tech Announce Co-Innovation Partnership to Advance Scalable AI in Supply Chain Orchestration admin Wed, 04/02/2025 - 11:27

Kinaxis, a global leader in supply chain orchestration, and the NSF AI Institute for Advances in Optimization (AI4OPT) at Georgia Tech today announced a new co-innovation partnership. This partnership will focus on developing scalable artificial intelligence (AI) and optimization solutions to address the growing complexity of global supply chains. AI4OPT operates under Tech AI, Georgia Tech’s AI hub, bringing together interdisciplinary expertise to advance real-world AI applications.

This particular collaboration builds on a multi-year relationship between Kinaxis and Georgia Tech, strengthening their shared commitment to turn academic innovation into real-world supply chain impact. The collaboration will span joint research, real-world applications, thought leadership, guest lectures, and student internships.

“In collaboration with AI4OPT, Kinaxis is exploring how the fusion of machine learning and optimization may bring a step change in capabilities for the next generation of supply chain management systems,” said Pascal Van Hentenryck, the A. Russell Chandler III Chair and professor at Georgia Tech, and director of AI4OPT and Tech AI at Georgia Tech.

Kinaxis’ AI-infused supply chain orchestration platform, Maestro™, combines proprietary technologies and techniques to deliver real-time transparency, agility, and decision-making across the entire supply chain — from multi-year strategic orchestration to last-mile delivery. As global supply chains face increasing disruptions from tariffs, pandemics, extreme weather, and geopolitical events, the Kinaxis–AI4OPT partnership will focus on developing AI-driven strategies to enhance companies’ responsiveness and resilience.

“At Kinaxis, we recognize the vital role that academic research plays in shaping the future of supply chain orchestration,” said Chief Technology Officer Gelu Ticala. “By partnering with world-class institutions like Georgia Tech, we’re closing the gap between AI innovation and implementation, bringing cutting-edge ideas into practice to solve the industry’s most pressing challenges.”

With more than 40 years of supply chain leadership, Kinaxis supports some of the world’s most complex industries, including high-tech, life sciences, industrial, mobility, consumer products, chemical, and oil and gas. Its customers include Unilever, P&G, Ford, Subaru, Lockheed Martin, Raytheon, Ipsen, and Santen.

About Kinaxis
Kinaxis is a global leader in modern supply chain orchestration, powering complex global supply chains and supporting the people who manage them, in service of humanity. Our powerful, AI-infused supply chain orchestration platform, Maestro™, combines proprietary technologies and techniques that provide full transparency and agility across the entire supply chain — from multi-year strategic planning to last-mile delivery. We are trusted by renowned global brands to provide the agility and predictability needed to navigate today’s volatility and disruption. For more news and information, please visit kinaxis.com or follow us on LinkedIn.

About AI4OPT
The NSF AI Institute for Advances in Optimization (AI4OPT) is one of the 27 National Artificial Intelligence Research Institutes set up by the National Science Foundation to conduct use-inspired research and realize the potential of AI. The AI Institute for Advances in Optimization (AI4OPT) is focused on AI for Engineering and is conducting cutting-edge research at the intersection of learning, optimization, and generative AI to transform decision making at massive scales, driven by applications in supply chains, energy systems, chip design and manufacturing, and sustainable food systems. AI4OPT brings together over 80 faculty and students from Georgia Tech, UC Berkeley, University of Southern California, UC San Diego, Clark Atlanta University, and the University of Texas at Arlington, working together with industrial partners that include Intel, Google, UPS, Ryder, Keysight, Southern Company, and Los Alamos National Laboratory. To learn more, visit ai4opt.org.

About Tech AI
Tech AI is Georgia Tech's hub for artificial intelligence research, education, and responsible deployment. With over $120 million in active AI research funding, including more than $60 million in NSF support for five AI Research Institutes, Tech AI drives innovation through cutting-edge research, industry partnerships, and real-world applications. With over 370 papers published at top AI conferences and workshops, Tech AI is a leader in advancing AI-driven engineering, mobility, and enterprise solutions. Through strategic collaborations, Tech AI bridges the gap between AI research and industry, optimizing supply chains, enhancing cybersecurity, advancing autonomous systems, and transforming healthcare and manufacturing. Committed to workforce development, Tech AI provides AI education across all levels, from K-12 outreach to undergraduate and graduate programs, as well as specialized certifications. These initiatives equip students with hands-on experience, industry exposure, and the technical expertise needed to lead in AI-driven industries. Bringing AI to the world through innovation, collaboration, and partnerships. Visit tech.ai.gatech.edu.

Subtitle

Joint Effort Targets Faster, Smarter Decision-Making Across Global Supply Chains

Summary sentence

Kinaxis and NSF AI4OPT at Georgia Tech expand their partnership to transform academic innovation into real-world supply chain impact through research, applications, thought leadership, lectures, and internships.

Summary

This collaboration between Kinaxis NSF AI4OPT at Georgia Tech builds on a multi-year relationship between Kinaxis and Georgia Tech, strengthening their shared commitment to turn academic innovation into real-world supply chain impact. The collaboration will span joint research, real-world applications, thought leadership, guest lectures, and student internships.

Dateline

Wed, 04/02/2025 - 12:00

Contact

Angela Barajas Prendiville | Director of Media Relations
aprendiville@gatech.edu

Associated importer

Keywords

go-research

go-researchnews

go-ai

News room topics

Campus and Community

Categories

Institute and Campus

Research

Mercury ID

681512

Source updated

Wed, 04/02/2025 - 11:27

Tech AI Fest 2025: Georgia Tech's Premier AI Event

admin — Mon, 24 Mar 2025 16:17:49 +0000

Tech AI Fest 2025: Georgia Tech's Premier AI Event admin Mon, 03/24/2025 - 12:17

March 20, 2025 – Tech AI, the AI hub at Georgia Tech, will host Tech AI Fest 2025 from March 26 to 28 at the Historic Academy of Medicine in Atlanta. This major AI event will bring together experts, researchers, industry professionals, policymakers, and students to explore the latest advancements and applications of artificial intelligence.

A Hub for AI Conversations

Tech AI Fest 2025 will bring together over 40 experts from top institutions and leading companies, including:

Academic Institutions: Georgia Tech, Carnegie Mellon University, Harvard University, the University of Texas at Austin, Washington State University, and Columbia University.
Leading Companies: J.P. Morgan, NVIDIA, Juniper Networks, Microsoft, OpenAI, Bosch USA, Kinaxis, MindsDB, Verdant Technologies, and Dandelion Science Corp.

Event Highlights

Days 1 – 2 (March 26 – 27) will focus on how AI is being used in industry, government, and research. Sessions will discuss AI's role in technological progress, economic growth, and policy development.
Day 3 (March 28) will showcase Georgia Tech's AI projects, including groundbreaking research, student presentations, and the launch of the Tech AI Alumni Group to strengthen industry-academic connections.

Commitment to AI Progress

Tech AI Fest 2025 shows Georgia Tech's dedication to advancing AI research, fostering innovation, and building meaningful partnerships across different sectors. The event is a platform for sharing knowledge, networking, and collaboration, putting attendees at the forefront of the AI revolution.

“AI is more than just an academic pursuit; it’s a force that’s reshaping industries, redefining education, and creating solutions for some of the biggest challenges we face today,” said Pascal Van Hentenryck, the A. Russell Chandler III Chair and professor in Georgia Tech’s H. Milton Stewart School of Industrial and Systems Engineering, director of the NSF Artificial Intelligence Institute for Advances in Optimization (AI4OPT), and director of Tech AI.

Registration and Participation

General admission is $25, but Georgia Tech students can attend for free while seats are available. Due to high demand, registration has reached capacity. Interested individuals can join the waiting list by contacting Josh Tullis at josh@corporate.gatech.edu.

For more information, visit the official event page.

About Tech AI

Tech AI is Georgia Tech’s hub for artificial intelligence research, education, and responsible deployment. It drives real-world AI solutions through innovation, collaboration, and industry partnerships. With over $120 million in active AI research funding and more than 370 publications at top AI conferences, Tech AI is ranked No. 5 in the U.S. for AI research and leads the way in AI-driven engineering and applied research. Supported by over $60 million in National Science Foundation funding, Tech AI plays a critical role in developing the next generation of AI leaders through world-class degree programs and specialized certifications. By bridging the gap between research and real-world applications, Tech AI collaborates with industry and government partners to optimize supply chains, enhance cybersecurity, advance autonomous systems, revolutionize healthcare, improve energy efficiency, and drive sustainable manufacturing.

Summary sentence

Georgia Tech's leading artificial intelligence event will bring together experts, researchers, industry professionals, policymakers, and students to explore the latest advancements and applications of AI.

Summary

Dateline

Fri, 03/21/2025 - 12:00

media@gatech.edu

Contact

Georgia Tech Media Relations

Keywords

go-research

go-researchnews

go-researchevents

go-ai

News room topics

Science and Technology

Mercury ID

681285

Source updated

Mon, 03/24/2025 - 12:16

Scientists Find Vehicles Susceptible to Remote Cyberattacks in Award-Winning Paper

dgivens8 — Mon, 02 Dec 2024 20:37:41 +0000

Scientists Find Vehicles Susceptible to Remote Cyberattacks in Award-Winning Paper dgivens8 Mon, 12/02/2024 - 15:37

Cybersecurity researchers have discovered new vulnerabilities that could provide criminals with wireless access to the computer systems in automobiles, aircraft, factories, and other cyber-physical systems.

The computers used in vehicles and other cyber-physical systems rely on a specialized internal network to communicate commands between electronics. Because it took place internally, it was traditionally assumed that attackers could only influence this network through physical access.

In collaboration with Hyundai, researchers from Georgia Tech’s Cyber-Physical Systems Security Research Lab (CPSec) observed that threat models used to evaluate the security of these technologies were outdated.

The team, led by Ph.D. student Zhaozhou Tang, found that vehicle technology advancements allowed attackers to launch new attacks, improve existing attacks, and circumvent current defense systems.

For example, Tang’s findings included the possibility for attackers to remotely compromise the computers used in cars and aircraft through Wi-Fi, cellular, Bluetooth, and other wireless channels.

“Our job was to thoroughly review existing information and find ways to protect against these attacks,” he said. “We found new threats and proposed a defense system that can protect against the new and old attacks.”

In response to their findings, the team developed ERACAN, the first comprehensive defense system against this new generation of attackers. Designed to detect new and old attacks, ERACAN can deploy defenses when necessary.

The system also classifies the attacks it reacts to, providing security experts with the tools for detailed analysis. It has a detection rate of 100% for all attacks launched by conventional methods and detects enhanced threat models 99.7% of the time.

The project received a distinguished paper award at the 2024 ACM Conference on Computer and Communications Security (CCS 24) held in Salt Lake City. Tang presented the paper at the October conference.

“This was Zhaozhou’s first paper in his Ph.D. program, and he deserves recognition for his groundbreaking work on automotive cybersecurity,” said Saman Zonouz, associate professor in the School of Cybersecurity and Privacy and the School of Electrical and Computer Engineering.

The U.S. Department of Homeland Security has designated the transportation sector as one of the nation’s 16 critical infrastructure sectors. Ensuring its security is vital to national security and public safety.

“Modern vehicles, which rely heavily on controller area networks for essential operations, are integral components of this infrastructure,” said Zonouz. “With the increasing sophistication of cyberthreats, safeguarding these systems has become critical to ensuring the resilience and security of transportation networks.”

This paper introduced to the scientific community the first comprehensive defense system to address advanced threats targeting vehicular controller area networks.

The CPSec team is putting the technology it has developed into practice in collaboration with Hyundai America Technical Center, Inc., which sponsors the work. Tang hopes ERACAN’s success will raise awareness of these new threats in the research community and industry.

“It will help them build future defenses,” he said. “We have demonstrated the best practice to defend against these attacks.”

Tang received his bachelor’s degree at Georgia Tech, where he first performed security-related work for the automobile industry. While working with Zonouz on his master’s degree, he decided to change course and pursue research initiatives like vehicle security in a Ph.D. program.

“It is interesting how it came full circle,” he said. “I will continue on this path of automobile security throughout my Ph.D.”

ERACAN: Defending Against an Emerging CAN Threat Model, was written by Zhaozhou Tang, Khaled Serag from the Qatar Computing Research Institute, Saman Zonouz, Berkay Celik and Dongyan Xu from Purdue University, and Raheem Beyah, professor and dean of the College of Engineering. The CPSec Lab is a collaboration between the School of Cybersecurity and Privacy and the School of Electrical and Computer Engineering.

Summary sentence

Georgia Tech researchers, with Hyundai, uncovered wireless vulnerabilities in vehicles' internal networks.

Summary

Georgia Tech researchers, with Hyundai, uncovered wireless vulnerabilities in vehicles' internal networks. They developed ERACAN, a defense system with near-perfect attack detection, addressing new and old threats. The project, recognized at CCS 2024, highlights the urgent need to secure critical transportation systems.

Dateline

Mon, 11/25/2024 - 12:00

jpopham3@gatech.edu

Contact

John Popham

Communications Officer II

School of Cybersecurity and Privacy

Associated importer

Keywords

Student

Research

Award

car

cars

Technology

go-researchnews

go-research

Daily Digest

College of Engineering; School of Electrical and Computer Engineering; College of Computing

News room topics

Science and Technology

Categories

Computer Science/Information Technology and Security

Engineering

Research

Student and Faculty

Student Competition WInners (academic, innovation, and research)

Student Research

Mercury ID

678727

Source updated

Mon, 12/02/2024 - 15:37

Minority English Dialects Vulnerable to Automatic Speech Recognition Inaccuracy

admin — Mon, 02 Dec 2024 16:41:41 +0000

Minority English Dialects Vulnerable to Automatic Speech Recognition Inaccuracy admin Mon, 12/02/2024 - 11:41

The Automatic Speech Recognition (ASR) models that power voice assistants like Amazon Alexa may have difficulty transcribing English speakers with minority dialects.

A study by Georgia Tech and Stanford researchers compared the transcribing performance of leading ASR models for people using Standard American English (SAE) and three minority dialects — African American Vernacular English (AAVE), Spanglish, and Chicano English.

Interactive Computing Ph.D. student Camille Harris is the lead author of a paper accepted into the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) this week in Miami.

Harris recruited people who spoke each dialect and had them read from a Spotify podcast dataset, which includes podcast audio and metadata. Harris then used three ASR models — wav2vec 2.0, HUBERT, and Whisper — to transcribe the audio and compare their performances.

For each model, Harris found SAE transcription significantly outperformed each minority dialect. The models more accurately transcribed men who spoke SAE than women who spoke SAE. Members who spoke Spanglish and Chicano English had the least accurate transcriptions out of the test groups.

While the models transcribed SAE-speaking women less accurately than their male counterparts, that did not hold true across minority dialects. Minority men had the most inaccurate transcriptions of all demographics in the study.

“I think people would expect if women generally perform worse and minority dialects perform worse, then the combination of the two must also perform worse,” Harris said. “That’s not what we observed.

“Sometimes minority dialect women performed better than Standard American English. We found a consistent pattern that men of color, particularly Black and Latino men, could be at the highest risk for these performance errors.”

Addressing underrepresentation

Harris said the cause of that outcome starts with the training data used to build these models. Model performance reflected the underrepresentation of minority dialects in the data sets.

AAVE performed best under the Whisper model, which Harris said had the most inclusive training data of minority dialects.

Harris also looked at whether her findings mirrored existing systems of oppression. Black men have high incarceration rates and are one of the people groups most targeted by police. Harris said there could be a correlation between that and the low rate of Black men enrolled in universities, which leads to less representation in technology spaces.

“Minority men performing worse than minority women doesn’t necessarily mean minority men are more oppressed,” she said. “They may be less represented than minority women in computing and the professional sector that develops these AI systems.”

Harris also had to be cautious of a few variables among AAVE, including code-switching and various regional subdialects.

Harris noted in her study there were cases of code-switching to SAE. Speakers who code-switched performed better than speakers who did not.

Harris also tried to include different regional speakers.

“It’s interesting from a linguistic and history perspective if you look at migration patterns of Black folks — perhaps people moving from a southern state to a northern state over time creates different linguistic variations,” she said. “There are also generational variations in that older Black Americans may speak differently from younger folks. I think the variation was well represented in our data. We wanted to be sure to include that for robustness.”

TikTok barriers

Harris said she built her study on a paper she authored that examined user-design barriers and biases faced by Black content creators on TikTok. She presented that paper at the Association of Computing Machinery’s (ACM) 2023 Conference on Computer Supported Cooperative Works.

Those content creators depended on TikTok for a significant portion of their income. When providing captions for videos grew in popularity, those creators noticed the ASR tool built into the app inaccurately transcribed them. That forced the creators to manually input their captions, while SAE speakers could use the ASR feature to their benefit.

“Minority users of these technologies will have to be more aware and keep in mind that they’ll probably have to do a lot more customization because things won’t be tailored to them,” Harris said.

Harris said there are ways that designers of ASR tools could work toward being more inclusive of minority dialects, but cultural challenges could arise.

“It could be difficult to collect more minority speech data, and you have to consider consent with that,” she said. “Developers need to be more community-engaged to think about the implications of their models and whether it’s something the community would find helpful.”

Subtitle

Summary sentence

The Automatic Speech Recognition (ASR) models that power voice assistants like Amazon Alexa may have difficulty transcribing English speakers with minority dialects.

Summary

Interactive Computing Ph.D. student Camille Harris is the lead author of a paper accepted into the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) this week in Miami.

Dateline

Fri, 11/15/2024 - 12:00

ndeen6@gatech.edu

Contact

Nathan Deen

Communications Officer

School of Interactive Computing

Associated importer

Keywords

speech recognition

bias

Research Horizons

go-research

go-researchnews

go-ai

Artifical Intelligence

inequality

News room topics

Society and Culture

Categories

Computer Science/Information Technology and Security

Mercury ID

678471

Source updated

Mon, 12/02/2024 - 11:39

go-research

Georgia Tech Celebrates 2025 Ph.D. Graduates in Cybersecurity and Privacy

Related links

Keywords

Categories

Despite Equal Publication Success in Climate Science, Women Leave the Field Earlier Than Men

Keywords

News room topics

National Report Urges FAA to Overhaul Air Traffic Controller Hiring and Training

Keywords

News room topics

Categories

Turning to CubeSats in the Search for Life Thousands of Light-Years from Earth

Keywords

Categories

SKYSCENES Dataset Could Lead to Safe, Reliable Autonomous Flying Vehicles

What's the Forecast?

Seeing the Whole Picture

Keywords

Categories

Rule the Pool This Summer and Make the Biggest Splash

Keywords

News room topics

Categories

Kinaxis and AI4OPT at Georgia Tech Announce Co-Innovation Partnership to Advance Scalable AI in Supply Chain Orchestration

Keywords

News room topics

Categories

Tech AI Fest 2025: Georgia Tech's Premier AI Event

Related links

Keywords

News room topics

Scientists Find Vehicles Susceptible to Remote Cyberattacks in Award-Winning Paper

Keywords

News room topics

Categories

Minority English Dialects Vulnerable to Automatic Speech Recognition Inaccuracy

Addressing underrepresentation

TikTok barriers

Keywords

News room topics

Categories