Enhancing Deep Learning with Bayesian Inference: Create more powerful, robust deep learning systems with Bayesian deep learning in Python - Softcover

Matt Benatan; Jochem Gietema; Marian Schneider

Softcover

ISBN 10: 180324688X ISBN 13: 9781803246888

Publisher: Packt Publishing, 2023

Synopsis

Develop Bayesian Deep Learning models to help make your own applications more robust.

Key Features

• Learn how advanced convolutions work
• Learn to implement a convolution neural network
• Learn advanced architectures using convolution neural networks
• Apply Bayesian NN to decrease weighted distribution

Book Description

Bayesian Deep Learning provides principled methods for developing deep learning models capable of producing uncertainty estimates.

Typical deep learning methods do not produce principled uncertainty estimates, i.e. they don’t know when they don’t know. Principled uncertainty estimates allow developers to handle unexpected scenarios in real-world applications, and therefore facilitate the development of safer, more robust systems.

Developers working with deep learning will be able to put their knowledge to work with this practical guide to Bayesian Deep Learning.

Learn building and understanding of how Bayesian Deep Learning can improve the way you work with models in production.

You’ll learn about the importance of uncertainty estimates in predictive tasks, and will be introduced to a variety of Bayesian Deep Learning approaches used to produce principled uncertainty estimates. You will be guided through the implementation of these approaches, and will learn how to select and apply Bayesian Deep Learning methods to real-world applications.

By the end of the book you will have a good understanding of Bayesian Deep Learning and the advantages it has to offer, and will be able to develop Bayesian Deep Learning models to help make your own applications more robust.

What you will learn

• Understanding the fundamentals of Bayesian Neural Networks
• Understanding the tradeoffs between different key BNN implementations/approximations
• Understanding the advantages of probabilistic DNNs in production contexts
• Knowing how to implement a variety of BDL methods, and how to apply these to real-world problems
• Understanding how to evaluate BDL methods and choose the best method for a given task

Who This Book Is For

Researchers and developers are looking for ways to develop more robust deep learning models through probabilistic deep learning.

The reader will know the fundamentals of machine learning, and have some experience of working with machine learning and deep learning models.

Table of Contents

1. Bayesian Inference in the Age of Deep Learning
2. Fundamentals of Bayesian Inference
3. Fundamentals of Deep Learning
4. Introducing Bayesian Deep Learning
5. Principled Approaches for Bayesian Deep Learning
6. Using the Standard Toolbox for Bayesian Deep Learning
7. Practical considerations for Bayesian Deep Learning
8. Applying Bayesian Deep Learning
9. Next steps in Bayesian Deep Learning

About this title

About the Authors

Matt Benatan is a Principal Research Scientist at Sonos and a Simon Industrial Fellow at the University of Manchester. His work involves research in robust multimodal machine learning, uncertainty estimation, Bayesian optimization, and scalable Bayesian inference.

Jochem Gietema is an Applied Scientist at Onfido in London where he has developed and deployed several patented solutions related to anomaly detection, computer vision, and interactive data visualisation.

Marian Schneider is an applied scientist in machine learning. His work involves developing and deploying applications in computer vision, ranging from brain image segmentation and uncertainty estimation to smarter image capture on mobile devices.

From the Back Cover

Develop Bayesian Deep Learning models to help make your own applications more robust. Key Features:Gain insights into the limitations of typical neural networks Acquire the skill to cultivate neural networks capable of estimating uncertainty Discover how to leverage uncertainty to develop more robust machine learning systems Book Description: Deep learning is revolutionizing our lives, impacting content recommendations and playing a key role in mission- and safety-critical applications. Yet, typical deep learning methods lack awareness about uncertainty. Bayesian deep learning offers solutions based on approximate Bayesian inference, enhancing the robustness of deep learning systems by indicating how confident they are in their predictions. This book will guide you in incorporating model predictions within your applications with care. Starting with an introduction to the rapidly growing field of uncertainty-aware deep learning, you'll discover the importance of uncertainty estimation in robust machine learning systems. You'll then explore a variety of popular Bayesian deep learning methods and understand how to implement them through practical Python examples covering a range of application scenarios. By the end of this book, you'll embrace the power of Bayesian deep learning and unlock a new level of confidence in your models for safer, more robust deep learning systems. What You Will Learn:Discern the advantages and disadvantages of Bayesian inference and deep learning Become well-versed with the fundamentals of Bayesian Neural Networks Understand the differences between key BNN implementations and approximations Recognize the merits of probabilistic DNNs in production contexts Master the implementation of a variety of BDL methods in Python code Apply BDL methods to real-world problems Evaluate BDL methods and choose the most suitable approach for a given task Develop proficiency in dealing with unexpected data in deep learning applications Who this book is for: This book will cater to researchers and developers looking for ways to develop more robust deep learning models through probabilistic deep learning. You're expected to have a solid understanding of the fundamentals of machine learning and probability, along with prior experience working with machine learning and deep learning models.

Edition details

• Publisher: Packt Publishing
• Publication date: 2023
• Language: English
• ISBN 10: 180324688X
• ISBN 13: 9781803246888
• Binding: Paperback
• Number of pages: 249