The master algorithm: how the quest for the ultimate learning machine will remake our world


  • Waldo Hasperué School of Computer Science, National University of La Plata, Argentina


machine learning


Nowadays, “machine learning” is present in several aspects of the current world, internet advisors, advertisements and “smart” devices that seem to know what we need in a given moment. These are some examples of the problems solved by machine learning. This book presents the past, the present and the future of the different types of machine learning algorithms. At the beginning of the book, the author takes us to the first years of the computing science, where a programmer had to do absolutely everything by himself to make an algorithm do a certain task. As time passes, there appeared the first algorithms that were capable of programming themselves learning from the available data. The author presents what he himself calls the five “tribes” of machine learning, the essence that defends each one and the kind of problems that are able to solve without problems. With a great amount of simple examples, the author depicts which advantages and disadvantages of the “master” algorithms of each “tribes” are, saying that the problem that a tribe solves perfectly well, another one cannot do it, and the other way about. The author suggests to get the best out of each “tribe” and make a unique learning algorithm able to learn without caring about the problem: the master algorithm.


Download data is not yet available.




How to Cite

Hasperué, W. (2015). The master algorithm: how the quest for the ultimate learning machine will remake our world. Journal of Computer Science and Technology, 15(02), p. 157–158. Retrieved from