Explore

Probabilistic Deep Learning with TensorFlow 2

Approx. 53 hours to complete

Save Course

Go to Course

Course Summary

Probabilistic Deep Learning with TensorFlow 2 is a course that teaches students how to build probabilistic models with deep learning techniques using TensorFlow 2. Students will learn how to use TensorFlow Probability, a library for probabilistic reasoning and statistical analysis, to build models for regression, classification, and other applications.

Key Learning Points

Learn how to build probabilistic models using deep learning techniques with TensorFlow 2
Understand how to use TensorFlow Probability to build models for regression, classification, and other applications
Gain experience implementing Bayesian neural networks and variational autoencoders

Job Positions & Salaries of people who have taken this course might have

- USA: $112,000
- India: ₹1,150,000
- Spain: €45,000
- USA: $112,000
- India: ₹1,150,000
- Spain: €45,000
- USA: $96,000
- India: ₹900,000
- Spain: €35,000
- USA: $112,000
- India: ₹1,150,000
- Spain: €45,000
- USA: $96,000
- India: ₹900,000
- Spain: €35,000
- USA: $130,000
- India: ₹1,400,000
- Spain: €55,000

Learning Outcomes

Build probabilistic models using deep learning techniques with TensorFlow 2
Implement Bayesian neural networks and variational autoencoders
Apply TensorFlow Probability to build models for regression, classification, and other applications

Prerequisites or good to have knowledge before taking this course

Familiarity with Python and deep learning concepts
Experience with TensorFlow or other deep learning frameworks

Course Difficulty Level

Intermediate

Course Format

Online
Self-paced
Video Lectures

Similar Courses

Probabilistic Graphical Models
Bayesian Methods for Machine Learning

Related Education Paths

Related Books

Description

Welcome to this course on Probabilistic Deep Learning with TensorFlow!

This course builds on the foundational concepts and skills for TensorFlow taught in the first two courses in this specialisation, and focuses on the probabilistic approach to deep learning. This is an increasingly important area of deep learning that aims to quantify the noise and uncertainty that is often present in real world datasets. This is a crucial aspect when using deep learning models in applications such as autonomous vehicles or medical diagnoses; we need the model to know what it doesn't know. You will learn how to develop probabilistic models with TensorFlow, making particular use of the TensorFlow Probability library, which is designed to make it easy to combine probabilistic models with deep learning. As such, this course can also be viewed as an introduction to the TensorFlow Probability library. You will learn how probability distributions can be represented and incorporated into deep learning models in TensorFlow, including Bayesian neural networks, normalising flows and variational autoencoders. You will learn how to develop models for uncertainty quantification, as well as generative models that can create new samples similar to those in the dataset, such as images of celebrity faces. You will put concepts that you learn about into practice straight away in practical, hands-on coding tutorials, which you will be guided through by a graduate teaching assistant. In addition there is a series of automatically graded programming assignments for you to consolidate your skills. At the end of the course, you will bring many of the concepts together in a Capstone Project, where you will develop a variational autoencoder algorithm to produce a generative model of a synthetic image dataset that you will create yourself. This course follows on from the previous two courses in the specialisation, Getting Started with TensorFlow 2 and Customising Your Models with TensorFlow 2. The additional prerequisite knowledge required in order to be successful in this course is a solid foundation in probability and statistics. In particular, it is assumed that you are familiar with standard probability distributions, probability density functions, and concepts such as maximum likelihood estimation, change of variables formula for random variables, and the evidence lower bound (ELBO) used in variational inference.

Outline

TensorFlow Distributions
Welcome to Probabilistic Deep Learning with TensorFlow 2
Interview with Paige Bailey
The TensorFlow Probability library
Univariate distributions
[Coding tutorial] Univariate distributions
Multivariate distributions
[Coding tutorial] Multivariate distributions
The Independent distribution
[Coding tutorial] The Independent distribution
Sampling and log probs
[Coding tutorial] Sampling and log probs
Trainable distributions
[Coding tutorial] Trainable distributions
Wrap up and introduction to the programming assignment
About Imperial College & the team
How to be successful in this course
Grading policy
Additional readings & helpful references
[Knowledge check] Standard distributions

Probabilistic layers and Bayesian neural networks
Welcome to week 2 - Probabilistic layers and Bayesian neural networks
The need for uncertainty in deep learning models
The DistributionLambda layer
[Coding tutorial] The DistributionLambda layer
Probabilistic layers
[Coding tutorial] Probabilistic layers
The DenseVariational layer
[Coding tutorial] The DenseVariational layer
Reparameterization layers
[Coding tutorial] Reparameterization layers
Wrap up and introduction to the programming assignment
Sources of uncertainty

Bijectors and normalising flows
Welcome to week 3 - Bijectors and normalising flows
Interview with Doug Kelly
Bijectors
[Coding tutorial] Bijectors
The TransformedDistribution class
[Coding tutorial] The Transformed Distribution class
Subclassing bijectors
[Coding tutorial] Subclassing bijectors
Autoregressive flows
RealNVP
[Coding tutorial] Normalising flows
Wrap up and introduction to the programming assignment
Change of variables formula

Variational autoencoders
Welcome to week 4 - Variational autoencoders
Encoders and decoders
[Coding tutorial] Encoders and decoders
Minimising KL divergence
[Coding tutorial] Minimising KL divergence
Maximising the ELBO
[Coding tutorial] Maximising the ELBO
KL divergence layers
[Coding tutorial] KL divergence layers
Wrap up and introduction to the programming assignment
Variational autoencoders

Capstone Project
Welcome to the Capstone Project
Goodbye video

Summary of User Reviews

Probabilistic Deep Learning with TensorFlow2 is a highly rated course on Coursera that covers the basics of deep learning and probabilistic modeling. Many users appreciated the course structure and the clear explanations provided by the instructor.

Key Aspect Users Liked About This Course

The course structure and the clear explanations provided by the instructor were highly appreciated by many users.

Pros from User Reviews

The course covers a broad range of topics related to deep learning and probabilistic modeling
The instructor provides clear explanations and examples that are easy to follow
The course is well-structured with well-paced lectures and assignments
The course provides a good balance between theory and practical applications
The course provides a good foundation for those interested in pursuing further studies in deep learning and probabilistic modeling

Cons from User Reviews

Some users found the course to be too theoretical and not practical enough
Some users found the assignments to be too difficult and time-consuming
Some users found the course to be too focused on TensorFlow and not enough on other deep learning frameworks
Some users found the course to be too slow-paced and not challenging enough
Some users found the course to be too expensive compared to other similar courses

Recommended for you

Complete Machine Learning & Data Science Bootcamp 2021

Learn Data Science, Data Analysis, Machine Learning (Artificial Intelligence) and Python with Tensorflow, Pandas & more!  You will go from zero to mastery!...

Save Course

Unsupervised Deep Learning in Python

Autoencoders, Restricted Boltzmann Machines, Deep Neural Networks, t-SNE and PCA This course is the next logical step in my deep learning, data science, and machine learning series....

Save Course

Applied Machine Learning in Python

This course will introduce the learner to applied machine learning, focusing more on the techniques and methods than on the statistics behind these methods....

Save Course

Generative Deep Learning with TensorFlow

In this course, you will: a) Learn neural style transfer using transfer learning: extract the content of an image (eg. swan), and the style of a painting (eg....

Save Course