Machine Learning algorithms are on the rise. Every year new techniques are presented that outdate the current leading algorithms. Some of them are only little advances or combinations of existing algorithms and others are newly created and lead to astonishing progress. For most techniques exist already great articles that explain the theory behind it and some of them offer also an implementation with code and tutorial. None did yet offer an overview of the current leading algorithms, so the idea came up to present the best algorithms per task based on the results achieved (performance scores are used). Of course, there are many more tasks and not all tasks can be presented. I tried to select the most popular fields and tasks and hope this might help to get a better understanding. The metiers on which this article will lay a focus are Computer Vision, Natural Language Processing, Speech Recognition.

All the fields, tasks and some of the algorithms are presented in the article. If you are interested only in a subpart, skip the to the section you want to dive in.

#algorithms #data-science #cv #machine-learning #nlp

Overview: State-of-the-Art Machine Learning Algorithms per Discipline & per Task
1.90 GEEK