Anna Choromanska

Anna Choromanska is an Associate Professor in the Department of Electrical and Computer Engineering at the NYU Tandon School of Engineering. She is affiliated with multiple centers, including the NYU Center for Data Science and the NYU Center for Advanced Technology in Communications. Her research focuses on deep learning, particularly on understanding and improving the optimization and generalization of deep neural networks. She has received several awards, including the NSF CAREER Award, Alfred P. Sloan Fellowship, and IBM Faculty Award. Her work has been applied in industry, including by Facebook and Baidu, and she is a contributor to the open-source system Vowpal Wabbit. She also leads the Learning Systems Laboratory at NYU.

Anna Choromanska is a researcher at New York University specializing in machine learning. Her work focuses on understanding the optimization landscape of deep neural networks, particularly the loss surfaces of multilayer networks. She has explored methods such as entropy SGD and low-pass filtering SGD to improve optimization and recovery of flat optima. Her research also includes applications in autonomous driving, such as sensor modality fusion with CNNs and visualization techniques like Visualbackprop. Additionally, she has contributed to online learning, clustering, and structured adaptive computations for efficient machine learning. Her publications highlight the intersection of theoretical and applied machine learning, emphasizing robustness, scalability, and interpretability.