Professor Amir Leshem's research focuses on the reconstruction of non-linear dynamic systems operating over complex networks. His work addresses one of the central challenges in modern signal processing and network science: how to recover and predict the behavior of systems with interacting components when only partial or noisy data is available.

By leveraging advanced tools from information theory, optimization, and machine learning, Professor Leshem develops algorithms that infer hidden states, model dynamic interactions, and enable robust forecasting in domains ranging from communication networks to biological systems. His research has broad implications for understanding complex phenomena such as epidemic spreading, neural activity, and distributed computation.