Shira Faigenbaum-Golovin is an Assistant Research Professor at Duke University. She earned her Ph.D. in Applied Mathematics from Tel Aviv University in 2021. Her research spans non-parametric estimation in low and high dimensions, manifold denoising and reconstruction, shape space analysis, and the theory of deep learning, particularly exploring the functions neural networks can approximate. She also works on interdisciplinary challenges in Digital Humanities, including archaeology, art history, remote sensing, and evolutionary anthropology.

For over 15 years, Shira has been developing machine-learning methods—drawing from mathematics, statistics, image and signal processing, and computer vision—to analyze the geometry of scattered data and address long-standing scientific questions. Her work, published in PNAS and PLOS ONE, includes groundbreaking studies on literacy rates in 7-8th century BCE, in Iron Age Israel and Judah. She designed an efficient multispectral imaging system that led to the discovery of a previously unknown 2,600-year-old inscription and developed algorithms to identify scribes in these ancient texts. In art history, she contributed to the automatic spectral analysis to uncover the underdrawings in The Adoration of the Kings by Botticelli and Lippi in collaboration with the National Gallery of London. In evolutionary anthropology, she studies primate species evolution by modeling collections of teeth as a manifold and applying machine-learning analysis.