Group items tagged

The blind application of machine learning runs the risk of amplifying biases present in data. Such a danger is facing us with word embedding, a popular framework to represent text data as vectors which has been used in many machine learning and natural language processing tasks. We show that even word embeddings trained on Google News articles exhibit female/male gender stereotypes to a disturbing extent. This raises concerns because their widespread use, as we describe, often tends to amplify these biases. Geometrically, gender bias is first shown to be captured by a direction in the word embedding. Second, gender neutral words are shown to be linearly separable from gender definition words in the word embedding. Using these properties, we provide a methodology for modifying an embedding to remove gender stereotypes, such as the association between between the words receptionist and female, while maintaining desired associations such as between the words queen and female. We define metrics to quantify both direct and indirect gender biases in embeddings, and develop algorithms to "debias" the embedding. Using crowd-worker evaluation as well as standard benchmarks, we empirically demonstrate that our algorithms significantly reduce gender bias in embeddings while preserving the its useful properties such as the ability to cluster related concepts and to solve analogy tasks. The resulting embeddings can be used in applications without amplifying gender bias.

"A trove of internal documents sheds light on the algorithms that Facebook's censors use to differentiate between hate speech and legitimate political expression."

An algorithm deduced the sexuality of people on a dating site with up to 91% accuracy, raising tricky ethical questions

"Insurer's algorithm analyses social media usage to identify safe drivers in unprecedented use of customer data"

Search term brings back mainly results of black women, which some say is evidence of bias. But algorithms may just be reflecting the wider social landscape

Algorithms used in American criminal justice is found to be biased against blacks

"Nearly three years after the company was called out, it hasn't gone beyond a quick workaround"

"The big companies developing them show no interest in fixing the problem."

If you searched for "three white teenagers" on Google Images earlier this month, the result spat up shiny, happy people in droves - an R.E.M. song in JPG format. The images, mostly stock photos, displayed young Caucasian men and women laughing, holding sports equipment or caught whimsically mid-selfie. If you searched for "three black teenagers," the algorithm offered an array of mug shots.

"Facebook's collection of data makes it one of the most influential organisations in the world. Share Lab wanted to look "under the bonnet" at the tech giant's algorithms and connections to better understand the social structure and power relations within the company. "

"Self-host your online community and shape your experience at the most granular level: add and remove features, change behaviours and appearance, tune, swap or turn off algorithms. You are in total control."

"A recent patent from the company judges your own creditworthiness by your friends."

Facebook has blocked plans by an insurer to view young drivers' profiles to help set car insurance premiums.

Google came under fire this week after its new Photos app categorized photos in one of the most racist ways possible. On June 28th, computer programmer Jacky Alciné found that the feature kept tagging pictures of him and his girlfriend as "gorillas."

"A California man has filed a class action suit against Fitbit, alleging that the company's sleep tracking is inaccurate and constitutes false advertising. "

Virgin sends subscribers a letter with details of what they watched

Virgin media sends Matt Webb a summary of what he's watched. Creepy!