Buy Now AED 170.99 + VAT

Monthly Subscription Starting at AED 99 + VAT

Probabilistic Deep Learning with TensorFlow 2
  • Level Expert
  • Duration 53 hours
  • Course by Imperial College London
  • Offered by

About

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.

Modules

Introduction to the course

    • 3 Videos
    • 4 Readings
    • 1 Discussion
    • 1 Plugin
    • 1 Assignment
Show info about module content

1 Assignment

  • [Knowledge check] Standard distributions

1 Discussions

  • Introduce yourself

3 Videos

  • Welcome to Probabilistic Deep Learning with TensorFlow 2
  • Interview with Paige Bailey
  • The TensorFlow Probability library

4 Readings

  • About Imperial College & the team
  • How to be successful in this course
  • Grading policy
  • Additional readings & helpful references

Univariate distributions

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Univariate distributions

2 Videos

  • Univariate distributions
  • [Coding tutorial] Univariate distributions

Multivariate distributions

    • 2 Videos
    • 2 Lab
Show info about module content

2 Labs

  • [Coding tutorial] Multivariate distributions
  • [Reading] Multivariate Gaussian with full covariance

2 Videos

  • Multivariate distributions
  • [Coding tutorial] Multivariate distributions

The Independent distribution

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] The Independent distribution

2 Videos

  • The Independent distribution
  • [Coding tutorial] The Independent distribution

Broadcasting rules

    • 1 Lab
Show info about module content

1 Labs

  • [Reading] Broadcasting rules

Sampling and log probs

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Sampling and log probs

2 Videos

  • Sampling and log probs
  • [Coding tutorial] Sampling and log probs

Trainable distributions

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Trainable distributions

2 Videos

  • Trainable distributions
  • [Coding tutorial] Trainable distributions

Programming Assignment: Naive Bayes and logistic regression

    • 1 Videos
    • 1 Lab
    • 1 Programming
Show info about module content

1 Programming

  • Naive Bayes and logistic regression

1 Labs

  • Naive Bayes and logistic regression

1 Videos

  • Wrap up and introduction to the programming assignment

Introduction to the week

    • 2 Videos
    • 1 Assignment
Show info about module content

1 Assignment

  • Sources of uncertainty

2 Videos

  • Welcome to week 2 - Probabilistic layers and Bayesian neural networks
  • The need for uncertainty in deep learning models

Maximum likelihood estimation

    • 1 Lab
Show info about module content

1 Labs

  • [Reading] Maximum likelihood estimation

The DistributionLambda layer

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] The DistributionLambda layer

2 Videos

  • The DistributionLambda layer
  • [Coding tutorial] The DistributionLambda layer

Probabilistic layers

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Probabilistic layers

2 Videos

  • Probabilistic layers
  • [Coding tutorial] Probabilistic layers

Bayes by backprop

    • 1 Lab
Show info about module content

1 Labs

  • [Reading] Bayes by backprop

The DenseVariational layer

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] The DenseVariational layer

2 Videos

  • The DenseVariational layer
  • [Coding tutorial] The DenseVariational layer

Reparameterization layers

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Reparameterization layers

2 Videos

  • Reparameterization layers
  • [Coding tutorial] Reparameterization layers

Programming Assignment: Bayesian convolutional neural network

    • 1 Videos
    • 1 Lab
    • 1 Programming
Show info about module content

1 Programming

  • Bayesian convolutional neural network

1 Labs

  • Bayesian convolutional neural network

1 Videos

  • Wrap up and introduction to the programming assignment

Introduction to the week

    • 2 Videos
    • 1 Lab
    • 1 Assignment
Show info about module content

1 Assignment

  • Change of variables formula

1 Labs

  • [Reading] Change of variables formula

2 Videos

  • Welcome to week 3 - Bijectors and normalising flows
  • Interview with Doug Kelly

Bijectors

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Bijectors

2 Videos

  • Bijectors
  • [Coding tutorial] Bijectors

Scale bijectors and LinearOperator

    • 1 Lab
Show info about module content

1 Labs

  • [Reading] Scale bijectors and LinearOperator

The TransformedDistribution class

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] The Transformed Distribution class

2 Videos

  • The TransformedDistribution class
  • [Coding tutorial] The Transformed Distribution class

Subclassing bijectors

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Subclassing bijectors

2 Videos

  • Subclassing bijectors
  • [Coding tutorial] Subclassing bijectors

Normalising flows

    • 3 Videos
    • 2 Lab
Show info about module content

2 Labs

  • [Reading] Autoregressive flows and RealNVP
  • [Coding tutorial] Normalising flows

3 Videos

  • Autoregressive flows
  • RealNVP
  • [Coding tutorial] Normalising flows

Programming Assignment: RealNVP

    • 1 Videos
    • 1 Lab
    • 1 Programming
Show info about module content

1 Programming

  • RealNVP

1 Labs

  • RealNVP

1 Videos

  • Wrap up and introduction to the programming assignment

Introduction to the week

    • 1 Videos
    • 1 Lab
    • 1 Assignment
Show info about module content

1 Assignment

  • Variational autoencoders

1 Labs

  • [Reading] Variational autoencoders

1 Videos

  • Welcome to week 4 - Variational autoencoders

Encoders and decoders

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Encoders and decoders

2 Videos

  • Encoders and decoders
  • [Coding tutorial] Encoders and decoders

Kullback-Leibler divergence

    • 2 Videos
    • 3 Lab
Show info about module content

3 Labs

  • [Reading] Kullback-Leibler divergence
  • [Coding tutorial] Minimising KL divergence
  • [Reading] Full covariance Gaussian approximation

2 Videos

  • Minimising KL divergence
  • [Coding tutorial] Minimising KL divergence

Maximising the ELBO

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] Maximising the ELBO

2 Videos

  • Maximising the ELBO
  • [Coding tutorial] Maximising the ELBO

KL divergence layers

    • 2 Videos
    • 1 Lab
Show info about module content

1 Labs

  • [Coding tutorial] KL divergence layers

2 Videos

  • KL divergence layers
  • [Coding tutorial] KL divergence layers

Programming Assignment: Variational autoencoder for Celeb-A

    • 1 Videos
    • 1 Lab
    • 1 Programming
Show info about module content

1 Programming

  • Variational autoencoder for Celeb-A

1 Labs

  • Variational autoencoder for Celeb-A

1 Videos

  • Wrap up and introduction to the programming assignment

Probabilistic generative models

    • 2 Videos
    • 1 PeerReview
    • 1 Plugin
    • 1 Lab
Show info about module content

1 Peer Review

  • Capstone Project

1 Labs

  • Capstone Project

2 Videos

  • Welcome to the Capstone Project
  • Goodbye video

Auto Summary

Explore the crucial domain of probabilistic deep learning with TensorFlow 2 in this expert-level course by Coursera. Delve into models that quantify uncertainty in real-world datasets, essential for applications like autonomous vehicles and medical diagnoses. Learn to develop probabilistic models using TensorFlow Probability, including Bayesian neural networks, normalizing flows, and variational autoencoders. Engage in hands-on coding tutorials and graded assignments, culminating in a Capstone Project. Ideal for learners with a solid foundation in probability and statistics. Subscription options include Starter, Professional, and Paid plans.

Dr Kevin Webster
Instructor

Dr Kevin Webster