More bad news for Google's beleaguered spinoff Jigsaw, whose flagship project is "Perspective," a machine-learning system designed to catch and interdict harassment, hate-speech and other undesirable online speech.
From the start, Perspective has been plagued by problems, but the latest one is a doozy: University of Washington experts have found that Perspective misclassifies inoffensive writing as hate speech far more frequently when the author is Black.
Specifically, candidate texts written in African American English (AAE) are 1.5x more likely to be rated as offensive than texts written in "white-aligned English."
The authors do a pretty good job of pinpointing the cause: the people who hand-labeled the training data for the algorithm were themselves biased, and incorrectly, systematically misidentified AAE writing as offensive. And since machine learning models are no better than their training data (though they are often worse!), the bias in the data propagated through the model.
In other words, Garbage In, Garbage Out remains the iron law of computing and has not been repealed by the deployment of machine learning systems.
We analyze racial bias in widely-used corpora of annotated toxic language, establishing correlations between annotations of offensiveness and the African American English (AAE) dialect. We show that models trained on these corpora prop-agate these biases, as AAE tweets are twice as likely to be labelled offensive compared to others.Finally, we introduce dialect and race priming,two ways to reduce annotator bias by highlightingthe dialect of a tweet in the data annotation, and show that it significantly decreases the likelihood of AAE tweets being labelled as offensive. Wefind strong evidence that extra attention should be paid to the confounding effects of dialect so as to avoid unintended racial biases in hate speech detection.
The Risk of Racial Bias in Hate Speech Detection [Maarten Sap, Dallas Card, Saadia Gabriel, Yejin Choi and Noah A. Smith/University of Washington]
(via Naked Capitalism)
Geneva ("Genetic Evasion") is a project from the University of Maryland's Breakerspace ("a lab dedicated to scaling-up undergraduate research in computer and network security"); in a paper presented today at the ACM's Conference on Computer and Communications Security, a trio of Maryland researchers and a UC Berkeley colleague present their work on evolutionary algorithms as […]
Daniel Moghimi, Berk Sunar, Thomas Eisenbarth and Nadia Heninger have published TPM-FAIL: TPM meets Timing and Lattice Attacks, their Usenix security paper, which reveals a pair of timing attacks against trusted computing chips ("Trusted Computing Modules" or TPMs), the widely deployed cryptographic co-processors used for a variety of mission-critical secure computing tasks, from verifying software […]
Next week, the Association for Computing Machinery (ACM) will host its inaugural Symposium on Computer Science and Law, whose sponsors include Palantir, Peter Thiel's notorious surveillance-tech company, which just renewed a $49m contract with ICE to provide technological aid for ICE's ethnic cleansing program, which has included mass family separations and the deaths of children […]
Sous vide cooking: It sounds fancy, but it’s actually one of the easiest and most reliable ways to cook. It’s the reason why many restaurants are able to put out delicious dishes with a consistent flavor. All you need is the right equipment, and that hasn’t always been available to those outside the resto crowd. […]
The more you use your computer, the more it becomes possible for others to use it too. Where there are anti-virus systems, there are hackers looking for a way to get around them. That’s why it’s important to get software that doesn’t just passively scout for viruses in the background. The folks behind GlassWire have […]
Knowledge is power. It’s a cliché, but sometimes things turn into a cliché because they’re true. If you’re making your way through the world of business and entrepreneurship, it only makes sense to read about the insights of people who have climbed that ladder before you. Trouble is, the modern workday doesn’t leave a lot […]