Kirill Eremenko is a lifestyle entrepreneur with 3 years of experience in the space of education and 7 years of experience in Data Science. Kirill is passionate about delivering high-quality accessible education to every human on this planet. Kirill is the founder and CEO of SuperDataScience, an online educational portal for Data Scientists.
Hadelin de Ponteves is passionate about data science, artificial intelligence, and deep learning. Hadelin teaches on Udemy and other platforms with tens of thousands of students Worldwide. Hadelin holds an engineering masters degree with a specialization in data science. He has also worked at Google where he implemented machine learning models for business analytics.
Presenters: Kirill Eremenko & Hadelin de Ponteves, Founder and CEO at SuperDataScience and Data Science Instructor at Udemy
Learn the essentials of machine learning including how Support Vector Machine, Naive Bayesian Classifier, and Upper Confidence Bound algorithms work. After this talk, you will have an intuitive understanding of these three three algorithms. And actual real life problems where they can be applied.
- Learn the intuition and how the Support Vector Machine (SVM) algorithm works.
- Learn the intuition and how the Naive Bayesian Classifier works.
- Learn the intuition behind and how the Upper Confidence Bound algorithm works, a Reinforcement Learning algorithm.
0:00 - 2:30: Introduction and Overview
2:30 - 6:24 Definitions
- Distinctions between AI, ML and DL.
- Types of problems in ML like regression, classification, clustering, association rules, reinforcement learning, natural language processing, deep learning.
6:24 - 20:47 Support Vector Machine (SVM)
- The intuition behind the SVM. Comparison with other algorithms.
- To cut between 11:32 to 12:51 there is a problem with the screen.
20:47 - 26:14 Bayes Theorem
26:14 - 41:52 Naive Bayes Classifier
- Intuition behind the algorithm and example of application.
41:52 - 1:04:38 Example of Reinforcement Learning
- Multi-Armed Bandit Problem
- Intuition behind Upper Confidence Bound (UCB) algorithm.
- Practical example in Spyder.
1:04:38 - 1:12:28 Questions