5 Tribes of AI https medium com42 aithe5

5 Tribes of AI https: //medium. com/42 ai/the-5 -tribes-of-the-ml-world-670 ebce 96 b 4 c

5 Tribes of AI The purpose of Machine Learning, according to the author, is to create The Master Algorithm: an algorithm capable of finding knowledge and generalizing from any kind of data. The algorithm must use paradigms and techniques from each and every tribe. “One Algorithm to rule them all, One algorithm to find them, One Algorithm to bring them all and in the darkness bind them, In the Land of Learning where the Data lies” https: //medium. com/42 ai/the-5 -tribes-of-the-ml-world-670 ebce 96 b 4 c

5 Tribes of AI Five basic methods of computer knowledge acquisition: • • • filling in gaps in existing knowledge mimicking the human brain simulating the evolutionary process reducing uncertainties making contrast between old and new sets of information https: //medium. com/42 ai/the-5 -tribes-of-the-ml-world-670 ebce 96 b 4 c

Pedro Domingos https: //youtu. be /B 8 J 4 uef. CQMc http: //homes. cs. washington. edu/~pedrod/smlr. pptx

Five Tribe Examples • Machine Learning Lingo • The Symbolists • Propositional logic • Deduction and induction • Decision trees • Knowledge engineering legacy • The Symbolists — Conclusion • The Connectionists • Neurological basis • Learning weights • Deep Learning • The Connectionists — Conclusion • The Evolutionaries • Nature's Algorithm • Genetic Processing • Case: Fringeling • Genetic Programming • Nature vs. Nurture • The Evolutionaries — Conclusion The Bayesians Bayes' Theorem Naïve Bayes Markov Models Bayesian Networks Bayesian learning Bayesians vs. Symbolists Case: Gen The Bayesians — Conclusion The Analogizers Nearest-neighbour algorithm Trouble in high dimensions Support Vector Machines Essence of Analogy The Analogizers — Conclusion https: //notes. fringeling. com/ On. The. Five. Tribes. Of. Machine. L earning/

Representation What do the models (solutions) look like? https: //notes. fringeling. com/On. The. Five. Tribes Of. Machine. Learning/ https: //medium. com/coinmonks/theartificial-neural-networks-handbook-part 1 -f 9 ceb 0 e 376 b 4

Evaluation How do we know if the models (solutions) are any good? Accuracy Squared Error

Optimisation How does the algorithm know if its models (solutions) are any good? Inverse deduction Gradient descent