Early Life and Education: From Lisbon to Academic Pursuits
Pedro Domingos was born in Lisbon, Portugal, in 1965. While details about his early life are limited, it is known that he earned a bachelor's degree in Electrical Engineering and Computer Science from Instituto Superior TĂ©cnico (IST), part of the University of Lisbon, graduating in 1988. His undergraduate education focused on electrical engineering and computer science, laying a strong technical foundation for his future research in machine learning.
He then pursued further studies in the United States at the University of California, Irvine (UCI), where he earned a Master’s degree in Computer Science in 1992 and a Ph.D. in Information and Computer Science in 1997. His doctoral dissertation, titled “Probabilistic Inference Based on Logic”, explored the integration of logical reasoning with probabilistic models, which later became the basis for his development of Markov Logic Networks.
After completing his Ph.D., Domingos returned to Portugal and served as an assistant professor at IST for two years (1997–1999). In 1999, he joined the University of Washington’s Department of Computer Science & Engineering, where he became an associate professor in 2008 and was promoted to full professor in 2012. His academic career reflects a balance between theoretical and applied research, particularly in the field of machine learning.
Career and Research: A Unifier in Machine Learning
Domingos' career has been centered around machine learning, with a particular focus on statistical relational learning. He is best known for developing Markov Logic Networks (MLNs), a framework introduced in 2003 in collaboration with Matt Richardson. MLNs combine first-order logic with probabilistic graphical models, enabling the handling of uncertainty in relational data. This approach has been widely applied in information extraction, natural language processing, social network analysis, and other domains.
In 2015, he published The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World, a book that explores five main paradigms in machine learning: Symbolists, Connectionists, Evolutionaries, Bayesians, and Analogizers. Domingos proposed a vision for unifying these approaches into a single master algorithm. The book was highly influential and was recommended by Bill Gates as a must-read on artificial intelligence and machine learning, demonstrating its impact on both academia and the general public.
His research also spans data stream analysis, cost-sensitive classification, adversarial learning, and other areas. In recognition of his foundational contributions, he received the ACM SIGKDD Innovation Award in 2014. He was elected a Fellow of AAAI in 2010 and received the IJCAI Distinguished Paper Award in 2014. Additionally, in 2018, he joined hedge fund D. E. Shaw & Co. to lead its machine learning research group but left the firm in 2019.
An unexpected detail about Domingos is his early interest in science fiction writing. During the 1990s at UC Irvine, he attended the Clarion West Writers Workshop, where he experimented with writing science fiction. However, he ultimately chose to focus on science, believing that technological advancements were outpacing the imagination of science fiction.
Awards and Honors
- 2014 ACM SIGKDD Innovation Award
- Elected AAAI Fellow in 2010
- 2014 IJCAI Distinguished Paper Award
- Sloan Research Fellowship, NSF CAREER Award, Fulbright Scholarship, and IBM Faculty Award
As of March 2025, his work has been cited over 61,000 times, underscoring his significant impact on the field of machine learning.
