Borna is from Tehran and studied Electrical Engineering at Sharif University of Technology, with a minor in Physics. During his undergraduate studies, he worked on machine learning, focusing on optimization and robustness. His research included aligning embedding models, stable and efficient convolutional architectures developed during a summer at EPFL, and adversarially robust embedding models for his bachelor’s thesis. He also explored accelerated sampling methods for Gaussian graphical models. Borna competed in the International Physics Olympiad, where he won a Silver Medal.
His coursework and background in physics sparked an interest in generative modelling and sampling. At StatML, Borna is excited to explore probabilistic machine learning and generative modelling, particularly diffusion and flow models, as well as steering and aligning large models, including language models.
