Immediate Download The Ethical Algorithm The Science of Socially Aware Algorithm Design 6th Edition Kearns Ebooks 2024
Immediate Download The Ethical Algorithm The Science of Socially Aware Algorithm Design 6th Edition Kearns Ebooks 2024
com
https://textbookfull.com/product/the-ethical-
algorithm-the-science-of-socially-aware-algorithm-
design-6th-edition-kearns/
https://textbookfull.com/product/the-algorithm-design-manual-3rd-
edition-steven-s-skiena/
textbookfull.com
https://textbookfull.com/product/algorithm-design-with-haskell-
richard-s-bird/
textbookfull.com
https://textbookfull.com/product/the-everyday-life-of-an-algorithm-
daniel-neyland/
textbookfull.com
https://textbookfull.com/product/sensing-and-control-for-autonomous-
vehicles-applications-to-land-water-and-air-vehicles-1st-edition-thor-
i-fossen/
textbookfull.com
Medical education at a glance First Edition Forrest
https://textbookfull.com/product/medical-education-at-a-glance-first-
edition-forrest/
textbookfull.com
https://textbookfull.com/product/popular-opinion-in-the-middle-ages-
channeling-public-ideas-and-attitudes-charles-w-connell/
textbookfull.com
Resilience to Climate Change Candice Howarth
https://textbookfull.com/product/resilience-to-climate-change-candice-
howarth/
textbookfull.com
THE ETHICAL ALGORITHM
MICHAEL KEARNS
AND
AARON ROTH
1 3 5 7 9 8 6 4 2
Printed by Sheridan Books, Inc. United States of America
Dedicated to our families
MK: Kim, Kate, and Gray
AR: Cathy, Eli, and Zelda
CONTENTS
Introduction
1 Algorithmic Privacy: From Anonymity to Noise
2 Algorithmic Fairness: From Parity to Pareto
3 Games People Play (With Algorithms)
4 Lost in the Garden: Led Astray by Data
5 Risky Business: Interpretability, Morality, and the
Singularity
Some Concluding Thoughts
Acknowledgments
References and Further Reading
Index
THE ETHICAL ALGORITHM
INTRODUCTION
ALGORITHMIC ANXIETY
We are allegedly living in a golden age of data. For practically any
question about people or society that you might be curious about,
there are colossal datasets that can be mined and analyzed to
provide answers with statistical certainty. How do the behaviors of
your friends influence what you watch on TV, or how you vote?
These questions can be answered with Facebook data, which records
the social network activity of billions of people worldwide. Are people
who exercise frequently less likely to habitually check their email?
For anyone who uses an Apple Watch, or an Android phone together
with the Google Fit app, the data can tell us. And if you are a retailer
who wants to better target your products to your customers by
knowing where and how they spend their days and nights, there are
dozens of companies vying to sell you this data.
Which all brings us to a conundrum. The insights we can get from
this unprecedented access to data can be a great thing: we can get
new understanding about how our society works, and improve public
health, municipal services, and consumer products. But as
individuals, we aren’t just the recipients of the fruits of this data
analysis: we are the data, and it is being used to make decisions
about us—sometimes very consequential decisions.
In December 2018, the New York Times obtained a commercial
dataset containing location information collected from phone apps
whose nominal purpose is to provide mundane things like weather
reports and restaurant recommendations. Such datasets contain
precise locations for hundreds of millions of individuals, each
updated hundreds (or even thousands) of times a day. Commercial
buyers of such data will generally be interested in aggregate
information—for example, a hedge fund might be interested in
tracking the number of people who shop at a particular chain of
retail outlets in order to predict quarterly revenues. But the data is
recorded by individual phones. It is superficially anonymous, without
names attached—but there is only so much anonymity you can
promise when recording a person’s every move.
For example, from this data the New York Times was able to
identify a forty-six-year-old math teacher named Lisa Magrin. She
was the only person who made the daily commute from her home in
upstate New York to the middle school where she works, fourteen
miles away. And once someone’s identity is uncovered in this way,
it’s possible to learn a lot more about them. The Times followed
Lisa’s data trail to Weight Watchers, to a dermatologist’s office, and
to her ex-boyfriend’s home. She found this disturbing and told the
Times why: “It’s the thought of people finding out those intimate
details that you don’t want people to know.” Just a couple of
decades ago, this level of intrusive surveillance would have required
a private investigator or a government agency; now it is simply the
by-product of widely available commercial datasets.
Clearly, we have entered a brave new world.
And it’s not only privacy that has become a concern as data
gathering and analysis proliferate. Because algorithms—those little
bits of machine code that increasingly mediate our behavior via our
phones and the Internet—aren’t simply analyzing the data that we
generate with our every move. They are also being used to actively
make decisions that affect our lives. When you apply for a credit
card, your application may never be examined by a human being.
Instead, an algorithm pulling in data about you (and perhaps also
about people “like you”) from many different sources might
automatically approve or deny your request. Though there are
benefits to knowing instantaneously whether your request is
approved, rather than waiting five to ten business days, this should
give us a moment of pause. In many states, algorithms based on
what is called machine learning are also used to inform bail, parole,
and criminal sentencing decisions. Algorithms are used to deploy
police officers across cities. They are being used to make decisions
in all sorts of domains that have direct and real impact on people’s
lives. All this raises questions not only of privacy but also of fairness,
as well as a variety of other basic social values including safety,
transparency, accountability, and even morality.
So if we are going to continue to generate and use huge datasets
to automate important decisions (a trend whose reversal seems
about as plausible as our returning to an agrarian society), we have
to think seriously about some weighty topics. These include limits on
the use of data and algorithms, and the corresponding laws,
regulations, and organizations that would determine and enforce
those limits. But we must also think seriously about addressing the
concerns scientifically—about what it might mean to encode ethical
principles directly into the design of the algorithms that are
increasingly woven into our daily lives. This book is about the
emerging science of ethical algorithm design, which tries to do
exactly that.
A BRIEF PREVIEW
In this book we will see how it is possible to expand the principles
on which machine learning is based to demand that they incorporate
—in a quantitative, measurable, verifiable manner—many of the
ethical values we care about as individuals and as a society.
Of course, the first challenge in asking an algorithm to be fair or
private is agreeing on what those words should mean in the first
place—and not in the way a lawyer or philosopher might describe
them, but in so precise a manner that they can be “explained” to a
machine. This will turn out to be both nontrivial and revealing, since
many of the first definitions we might consider turn out to have
serious flaws. In other cases we will see there may be several
intuitive definitions that are actually in conflict with each other.
But once we’ve settled on our definitions, we can try to internalize
them in the machine learning pipeline, encoding them into our
algorithms. But how? Machine learning already has a “goal,” which is
to maximize predictive accuracy. How do we introduce new goals
such as fairness and privacy into the code without “confusing” the
Random documents with unrelated
content Scribd suggests to you:
The Project Gutenberg eBook of Aucassin ja
Nicolette: Laulutarina
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Language: Finnish
AUCASSIN JA NICOLETTE
Laulutarina
Muinais-ranskasta suomentanut
Eino Palola
*****
Tähän itämaiseen tarinaryhmään kuuluu kertomus Aucassinista ja
Nicolettestakin, mutta ennen kuin ryhdymme sitä tarkemmin
käsittelemään, lienee syytä ottaa selville, kuinka kirjalliset tuotteet
siirrettiin yleisön nautittaviksi, kuinka ne levisivät ja kulkivat edelleen.
*****
Suomentaja.
AUCASSIN JA NICOLETTE
Nyt lauletaan:
Nyt lauletaan:
Nyt lauletaan:
Nyt lauletaan:
"Mielelläni, herra."
Nyt lauletaan: