Mansour is considered a worldwide leading researcher and his research spans a wide variety of different areas: machine learning, reinforcement learning, algorithmic game theory, and communication networks. Machine learning focuses on the ability to provide an automatic prediction, based on huge amounts of processed data, and is already a key technology in many applications and has the potential to greatly impact our everyday life. Reinforcement learning is aimed at interactive environments, and its goal is learning through interactions, with already established success stories exhibiting human level performance in a wide variety of tasks, and is projected to play a crucial role in robotics and autonomous driving. Algorithmic game theory bridges between economics and computer science, highlights issues of incentives (from the economics perspective) and computation (from the computer science perspective), and its importance stems from the rise of electronic commerce and the use of electronic auctions. Communication networks are the infrastructure on which the information revolution has materialized, and one cannot overestimate the importance of communication networks in establishing our connected global village. Those fields are the cornerstones of today’s information technologies, and will impact humanity in the years to come.

Mansour’s works have greatly influenced each of the aforementioned research fields, providing solid theoretical foundations and profound algorithmic methodologies to address fundamental challenges. His contributions also bridged between those fields, creating new and surprising connections. He has fostered computational learning theory from its infancy days, helped shape it and grow its influence on the greater machine learning community, and increase its importance until it became today an integral part of mainstream machine learning. His works on reinforcement learning established methodologies to handle large state Markov Decision Processes (MDPs), which received even greater importance today, with the numerous applications of deep reinforcement learning. His works on algorithmic game theory have introduced many concepts from learning theory and established strong connections between the two fields. He made early contributions to communication networks, forming an algorithmic perspective of scheduling and protocol design in communication networks.