Information Science, Machine Learning, and Data Analytics Strategies for Advertising and marketing, Digital Media, On-line Promoting, and More
What you’ll be taught
- Use adaptive algorithms to improve A/B testing performance
- Understand the difference between Bayesian and frequentist statistics
- Apply Bayesian methods to A/B testing
- Probability (joint, marginal, conditional distributions, steady and discrete random variables, PDF, PMF, CDF)
- Python coding with the Numpy stack
This course is all about A/B testing.
A/B testing is used everywhere. Advertising and marketing, retail, newsfeeds, internet advertising, and more.
A/B testing is all about comparing things.
If you’re an information scientist, and you need to tell the rest of the corporate, “emblem A is healthier than emblem B”, nicely you can’t simply say that without proving it utilizing numbers and statistics.
Conventional A/B testing has been around for a long time, and it’s full of approximations and complicated definitions.
On this course, while we’ll do traditional A/B testing in order to respect its complexity, what we’ll finally get to is the Bayesian machine learning means of doing issues.
First, we’ll see if we will enhance on traditional A/B testing with adaptive strategies. These all help you clear up the explore-exploit dilemma.
You’ll study concerning the epsilon-greedy algorithm, which you could have heard about within the context of reinforcement learning.
We’ll improve upon the epsilon-greedy algorithm with a similar algorithm called UCB1.
Lastly, we’ll improve on each of these by using a totally Bayesian strategy.
Why is the Bayesian method interesting to us in machine studying?
It’s an entirely different way of thinking about probability.
It’s a paradigm shift.
You’ll most likely want to come again to this course several times earlier than it absolutely sinks in.
It’s also powerful, and lots of machine studying experts typically make statements about how they “subscribe to the Bayesian college of thought”.
In sum – it’s going to give us a variety of powerful new tools that we will use in machine learning.
The stuff you’ll be taught on this course are usually not solely relevant to A/B testing, however slightly, we’re utilizing A/B testing as a concrete instance of how Bayesian methods might be utilized.
You’ll be taught these basic tools of the Bayesian methodology – by means of the instance of A/B testing – and then you definately’ll have the ability to carry these Bayesian methods to extra superior machine studying models sooner or later.
See you in class!
“In case you can’t implement it, you don’t perceive it”
- Or as the good physicist Richard Feynman mentioned: “What I can’t create, I don’t perceive”.
- My programs are the ONLY courses where you’ll discover ways to implement machine studying algorithms from scratch
- Different programs will train you how to plug in your data into a library, however do you actually need assist with three strains of code?
- After doing the same factor with 10 datasets, you notice you didn’t be taught 10 issues. You discovered 1 factor, and simply repeated the identical three strains of code 10 occasions…
- Chance (joint, marginal, conditional distributions, steady and discrete random variables, PDF, PMF, CDF)
- Python coding: if/else, loops, lists, dicts, units
- Numpy, Scipy, Matplotlib
WHAT ORDER SHOULD I TAKE YOUR COURSES IN?:
- Take a look at the lecture “Machine Studying and AI Prerequisite Roadmap” (obtainable within the FAQ of any of my programs, including the free Numpy course)
Who this course is for:
- Students and professionals with a technical background who need to be taught Bayesian machine studying methods to use to their knowledge science work
Bayesian Machine Studying in Python: A/B Testing Free Obtain