Group items tagged

A new world is coming. It's scary. Freaky. Over the freaky line, if you will. But it is coming. Investors like Ron Conway and Marc Andreessen are investing in it. Companies from Google to startups you've never heard of, like Wovyn or Highlight, are building it. With more than a couple of new ones already on the way that you'll hear about over the next six months.

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

"Project Tango is Google's attempt to build a device which is intimately acquainted with its surroundings "

"Project Tango is Google's attempt to build a device which is intimately acquainted with its surroundings "

"FACEBOOK. Instagram. Google. Twitter. All services we rely on - and all services we believe we don't have to pay for. Not with cash, anyway. But ad-financed Internet platforms aren't free, and the price they extract in terms of privacy and control is getting only costlier. A recent Pew Research Center poll shows that 93 percent of the public believes that "being in control of who can get information about them is important," and yet the amount of information we generate online has exploded and we seldom know where it all goes."

Someone who knows things about us has some measure of control over us, and someone who knows everything about us has a lot of control over us. Surveillance facilitates control.

"Amazon, Apple and Google all employ staff who listen to customer voice recordings from their smart speakers and voice assistant apps."

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.

Open source, privacy aware Google Home & Amazon Alexa iot device and much more

Researchers say Google's ad-targeting system sometimes makes troubling decisions based on data about gender and other personal characteristics.

Is your digital assistant taking orders behind your back? Scientists from China's Zheijiang University have proved it's possible, publishing new research that demonstrates how Siri, Alexa, and other voice-activated programs can be controlled using inaudible ultrasound commands.

Fair and secure google docs

"Schemer - The beginning of everything worth doing! "