New technologies enabled by AI are not only changing aspects of our daily lives, but also reshaping our notions of what is possible. From unlocking new possibilities in the classroom to advancing solutions to some of society's biggest challenges, we are in the midst of a machine learning revolution. As a talented collection of America's brightest young people, it's no surprise that many of this year's Regeneron Science Talent Search finalists are emerging professionals in their fields. Meet a few people who are exploring the potential of AI in research.
sophie chen
Caddo Parish Magnet High School (Shreveport, Louisiana)
Sophie's project aims to address the time-consuming and error-prone process of analyzing tissue during resection of squamous cell carcinoma. She developed a convolutional neural network machine learning model that reviews microscopic images of tissue that can recognize fundamental differences between tumors, inflammation, and benign areas. Regarding her future goals, Sophie said: Regardless of where we end up, we want to harness the power of AI to bring about transformative change in healthcare. ”
Zeyneb Kaya (5th place winner)
Saratoga High School (Saratoga, California)
Drawing on his deep passion for linguistics and computer science, Zeyneb developed a natural language processing algorithm similar to ChatGPT with the goal of preserving endangered languages. Many large-scale language models benefit from large amounts of readily available text for training, but less commonly spoken languages often do not have such an advantage. To address the relative lack of training materials, Zeyneb created an algorithm to generate additional text in these target languages, allowing subsequent natural language processing algorithms to work better.
alexis lee
Hamilton High School (Chandler, Arizona)
Alexis has developed BrainSTEAM, a suite of machine learning tools that identifies key neural patterns in neuropsychiatric diseases. BrainSTEAM works by analyzing MRI images to determine the strength of connections between brain regions. This can help identify key neural patterns in conditions such as autism spectrum disorders, ADHD, and depression. Alexis' research could help medical professionals diagnose these conditions earlier and more accurately, as current diagnoses often rely solely on subjective assessments.
Achyuta Rajaram (1st place winner)
Phillips Exeter Academy (Exeter, New Hampshire)
Achyuta has created automated methods to better understand how deep learning models make decisions. In particular, when it comes to computer vision problems such as how AI models detect objects and generate images, Achyuta argues that the ambiguity of what these models are “thinking” as they produce outputs from inputs We are working hard to answer these questions. His research has the potential to make widespread application of these models more effective, safe, and fair in the future.
Noh Geun Hyun (Kyle)
Bronx High School of Science (Bronx, New York)
Kyle has developed machine learning models that can save valuable time in the discovery and development of new drugs, especially for Alzheimer's disease. His model identifies potential substances and compounds that could more effectively inhibit gene expression of key proteins. In addition, Kyle founded the school's machine learning club and partners with local non-profit organizations to provide free computer science education to area middle school students.
Riya Chagi
Phillips Exeter Academy (Exeter, New Hampshire)
Riya used computer vision to identify how an AI model learns a patient's race and ethnicity from anonymous medical images. This understanding is key to minimizing bias in healthcare, especially as AI tools become more widespread across the medical field. Talking about her future goals, Riya says: “I would like to work not only as an AI ethicist, but also in the field of climate technology. We love the wildlife around us and think there is a lot to learn from it.”
