Dr. Noam Razin

Deep learning is experiencing unprecedented success, and is largely responsible for the technological breakthroughs referred to in the public as “artificial intelligence” (AI). However, despite its extreme popularity in science and industry, the formal understanding of deep learning is limited.

Noam’s research focuses on the theoretical and algorithmic foundations of deep learning. Through a combination of mathematical analysis and experimentation, he aims to develop theories that shed light on how neural networks work, as well as bring forth principled methods for improving their efficiency, reliability, and performance.

Noam completed his PhD in Computer Science at Tel Aviv University, where he was advised by Nadav Cohen. His doctoral thesis applies tools from dynamical systems theory and tensor analysis for explaining why neural networks perform well over natural data, such as audio, images, and text. Aside from strengthening the formal understanding of deep learning, the developed theory led to several practical data preprocessing and regularization algorithms.

Now as a postdoctoral fellow at Princeton Language and Intelligence, a part of Princeton University, Noam is working on fundamental questions regarding the finetuning and alignment of language models.

For his research, Noam received the Apple Scholars in AI/ML PhD fellowship, the Tel Aviv University Center for AI and Data Science excellence fellowship, and the Deutsch Prize for PhD candidates.