A UCF research team placed eighth in the 2025 EEG Challenge, a global machine learning competition that challenges participants to predict behavioral responses from brain data. The Knights, who call themselves Team Marque, fielded 8,400 applications, including applications from research institutions and technology companies such as Meta and Emotiv.
The winning team included Mubarak Shah, director of the UCF Institute for Artificial Intelligence (IAI); Helen Huang and Qiushi Fu, associate professors of biomedical engineering; Yue Wen, assistant professor of biomedical engineering; Abhilash Durgam is a PhD student working in the Computer Vision Research Center. and Jerry Fu, a postdoctoral fellow supervised by Huang and Wen.
As a top 10 winner, Team Marque’s code will be added to the contest’s open source repository, contributing to future advances in brainwave research. They will also receive a certificate in recognition of their achievements. Shah said being in the top 10 of the world’s best venues for AI and machine learning is a major accomplishment for UCF and its newly formed IAI.
“This speaks to the strength of UCF’s interdisciplinary culture,” Shah said.
“Our students and faculty combine expertise in machine learning, neuroscience, signal processing, and computer vision to compete against the best teams in the world.” — Mubarak Shah, Chair Professor
Contestants had to win two separate challenges that utilized data from the Healthy Brain Network, which includes the brainwaves of more than 3,000 multitasking children. Task 1 asked the team to improve the predicted reaction times of subjects who saw contrast changes in images, while task 2 asked them to improve their predictions of subjects’ mental health characteristics.
Durgum says the secret to Team Marque’s success was looking for a pattern that fits everyone.
“Rather than treating this as a regression problem to predict a number, we used a classification approach that teaches the model to recognize a person’s unique ‘profile’,” Durgum says. “This encouraged the model to understand distinct characteristics of the individual, rather than treating the task as a simple math problem.”
The team’s work is more than just an accomplishment for themselves and the university; scientists can now use the code to advance brain wave research.
“Our open source repository supports open science efforts, which I believe are necessary to make significant advances in brain wave research at a faster rate than any group could achieve alone,” Huang says. “Being able to predict the mental health characteristics of developing children is a difficult problem with major societal implications, but one that we can collectively solve as a field faster by working in parallel and sharing data and code so that groups don’t have to repeat what has already been tried.”
The Marche team was formed after Durgum reached out to Huang to learn more about brain waves. Each had already formed a team for the contest, but decided to join forces for a better result. For Huang, this tournament has a personal connection to him as one of the organizers. Seyed Yahya Shirazi ’21PhDa former student of hers.
“I don’t think I could have made it into the top 10 without putting in the effort,” Huang said. “Together, we were able to consider fundamentally different approaches in parallel, first identifying the most promising approaches, and then focusing on optimizing specific parameters.”
