Udemy
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Probabilistic Programming with Python and Julia
Rating: 3.3 out of 5(30 ratings)
216 students

Probabilistic Programming with Python and Julia

Introduction and simple examples to start into probabilistic programming
Created byBert Gollnick, Sebastian Kaus
Last updated 7/2019
English
English [Auto],

What you'll learn

  • Introduction to probabilistic programming
  • Bayesian statistics
  • Markov Chain Monte Carlo
  • Gaussian Mixture Models
  • Bayesian Logistic Regression
  • Bayesian Linear Regression

Course content

7 sections • 26 lectures • 2h 39m total length

  • Course Overview2:49

    Explore statistics basics, distributions, and monte carlo markov chains while setting up Python and Julia environments, and preview bayesian linear and logistic regression with credible intervals.

  • Bayesian Statistics11:45

    Explore Bayesian statistics and how it differs from classical statistics by incorporating prior beliefs, likelihood, and posterior updates using Bayes' formula, with density functions and real-world examples.

  • Distributions: Introduction1:05

    Explore the key distributions and their features, classify them as parametric or non parametric, and distinguish discrete versus continuous outcomes in probabilistic models.

  • Distributions: Uniform Distribution4:47

    Explore the uniform distribution, where all values between A and B are equally likely, shown via histograms and samples; compute the mean (A+B)/2 and study its linear CDF.

  • Distributions: Normal Distribution4:42

    Explain how the normal distribution is defined by mu and sigma, its mean and standard deviation, forming a bell-shaped curve.

  • Distributions: Binomial Distribution2:09

    Explore the binomial distribution, a discrete two-outcome model with mean equal to n p and variance calculated differently, illustrated by coin toss data and convergence to normal for large n.

  • Distributions: Poisson Distribution1:31

    The Poisson distribution models the probability of events in a time interval with average rate lambda and observed count k, with examples such as radioactive decay and calls per hour.

  • Monte Carlo Markov Chain8:10

    Learn how Markov chain Monte Carlo bridges prior and posterior distributions by sampling high-dimensional models, using Monte Carlo simulations and acceptance-based sampling methods such as Metropolis-Hastings.

Requirements

  • Python
  • Julia
  • Elementary understanding of statistics

Description

You want to know and to learn one of the top 10 most influencial algorithms of the 20th century? Then you are right in this course. We will cover many powerful techniques from the field of probabilistic programming. This field is fast-growing, because these technique are getting more and more famous and proof to be efficient and reliable.

We will cover all major fields of Probabilistic Programming: Distributions, Markov Chain Monte Carlo, Gaussian Mixture Models, Bayesian Linear Regression, Bayesian Logistic Regression, and hidden Markov models.

For each field, the algorithms are shown in detail: Their core concepts are presented in 101 lectures. Here, you will learn how the algorithm works. Then we implement it together in coding lectures. These are available for Python and Julia. With this knowledge you can clearly identify a problem at hand and develop a plan of attack to solve it.

Mastering this course will enable you to understand the concepts of probabilistic programming and you will be able to apply this in your private and professional projects.

Who this course is for:

  • Python and Julia users who like to learn probabilistic programming

Report abuse