Professor Pedro Domingos, Técnico alumnus and currently professor at the University of Washington, presented a lecture at Técnico on the topic “The Five Tribes of Machine Learning, and What You Can Take from Each”, within the INESC-ID Distinguished Lecture Series.

 
The professor – whose book, “The Master Algorithm”, was recommended by Bill Gates to understand Artificial Intelligence – spoke about the five “tribes” of Machine Learning, and what you can take from each. There are five main schools of thought in machine learning, and each has its own master algorithm, but “the truth is that a single Master Algorithm must combine the key features of all of them.”

 
According to professor Pedro Domingos, in order to create this single and universal algorithm, which will allow the development of personal robots and the cure for cancer, it will be necessary to think outside of the box. “Many ideas are still lacking in Machine Learning – I think people who don’t work in this area or students who have not yet their own thoughts about this subject will come up with such ideas easily.”

 
In the end, Pedro Domingos answered the questions from the audience and left an advice: “I usually teach my PhD students Machine Learning but more than that I teach them how to do research. What they research today may change within 10 or 20 years, but if they know how to do research, they use it for everything”.

 
“You must be prepared to discard what you have learned. I usually say ‘learn everything, but do not believe in anything‘, because your job is to make the old knowledge base obsolete.”