[Oreilly] Privacy-Preserving Machine Learning Free Download
O’Reilly members get limitless entry to books, stay occasions, programs curated by job position, and extra from O’Reilly and almost 200 prime publishers.
Book description
Keep delicate person information protected and safe with out sacrificing the efficiency and accuracy of your machine studying fashions.
In Privacy Preserving Machine Learning, you’ll be taught:
- Privacy issues in machine studying
- Differential privateness methods for machine studying
- Privacy-preserving artificial information era
- Privacy-enhancing applied sciences for information mining and database purposes
- Compressive privateness for machine studying
Privacy Preserving Machine Learning is a complete information to avoiding information breaches in your machine studying initiatives. You’ll familiarize yourself with fashionable privacy-enhancing methods reminiscent of differential privateness, compressive privateness, and artificial information era. Based on years of DARPA-funded cybersecurity analysis, ML engineers of all ability ranges will profit from incorporating these privacy-preserving practices into their mannequin improvement. By the time you’re accomplished studying, you’ll be capable to create machine studying techniques that protect person privateness with out sacrificing information high quality and mannequin efficiency.
About the Technology
Machine studying purposes want large quantities of knowledge. It’s as much as you to maintain the delicate info in these information units non-public and safe. Privacy preservation occurs at each level within the ML course of, from information assortment and ingestion to mannequin improvement and deployment. This sensible e-book teaches you the abilities you’ll must safe your information pipelines finish to finish.
About the Book
Privacy Preserving Machine Learning explores privateness preservation methods by way of real-world use instances in facial recognition, cloud information storage, and extra. You’ll find out about sensible implementations you may deploy now, future privateness challenges, and find out how to adapt current applied sciences to your wants. Your new abilities construct in the direction of an entire safety information platform venture you’ll develop within the remaining chapter.
What’s Inside
- Differential and compressive privateness methods
- Privacy for frequency or imply estimation, naive Bayes classifier, and deep studying
- Privacy-preserving artificial information era
- Enhanced privateness for information mining and database purposes
About the Reader
For machine studying engineers and builders. Examples in Python and Java.
About the Authors
J. Morris Chang is a professor on the University of South Florida. His analysis initiatives have been funded by DARPA and the DoD. Di Zhuang is a safety engineer at Snap Inc. G. Dumindu Samaraweera is an assistant analysis professor on the University of South Florida. The technical editor for this e-book, Wilko Henecka, is a senior software program engineer at Ambiata the place he builds privacy-preserving software program.
Quotes
An in depth remedy of differential privateness, artificial information era, and privacy-preserving machine-learning methods with related Python examples. Highly really useful!
– Abe Taha, Google
A beautiful synthesis of theoretical and sensible. This e-book fills an actual want.
– Stephen Oates, Allianz
The definitive supply for creating privacy-respecting machine studying techniques. This space in data-rich environments is so necessary to grasp!
– Mac Chambers, Roy Hobbs Diamond Enterprises
Covers all facets for information privateness, with good sensible examples.
– Vidhya Vinay, Streamingo Solutions
by Morris Chang, Dumindu Samaraweera, Di Zhuang
Released May 2023
Publisher(s): Manning Publications
ISBN: 9781617298042
Size: 1.15 GB
https://www.oreilly.com/library/view/privacy-preserving-machine-learning/9781617298042/.
