Civil Rights & Constitutional Law


Europeans asked Google for their 'Right to be Forgotten' 2.4 million times

Mashable

After three years in effect, the European ruling with a name that sounds like it's straight out of a science-fiction book is revealing the things people most want to hide about themselves online.


'Least Desirable'? How Racial Discrimination Plays Out In Online Dating

NPR

In 2014, user data on OkCupid showed that most men on the site rated black women as less attractive than women of other races and ethnicities. That resonated with Ari Curtis, 28, and inspired her blog, Least Desirable.


On the Hardness of Inventory Management with Censored Demand Data

arXiv.org Machine Learning

We consider a repeated newsvendor problem where the inventory manager has no prior information about the demand, and can access only censored/sales data. In analogy to multi-armed bandit problems, the manager needs to simultaneously "explore" and "exploit" with her inventory decisions, in order to minimize the cumulative cost. We make no probabilistic assumptions---importantly, independence or time stationarity---regarding the mechanism that creates the demand sequence. Our goal is to shed light on the hardness of the problem, and to develop policies that perform well with respect to the regret criterion, that is, the difference between the cumulative cost of a policy and that of the best fixed action/static inventory decision in hindsight, uniformly over all feasible demand sequences. We show that a simple randomized policy, termed the Exponentially Weighted Forecaster, combined with a carefully designed cost estimator, achieves optimal scaling of the expected regret (up to logarithmic factors) with respect to all three key primitives: the number of time periods, the number of inventory decisions available, and the demand support. Through this result, we derive an important insight: the benefit from "information stalking" as well as the cost of censoring are both negligible in this dynamic learning problem, at least with respect to the regret criterion. Furthermore, we modify the proposed policy in order to perform well in terms of the tracking regret, that is, using as benchmark the best sequence of inventory decisions that switches a limited number of times. Numerical experiments suggest that the proposed approach outperforms existing ones (that are tailored to, or facilitated by, time stationarity) on nonstationary demand models. Finally, we extend the proposed approach and its analysis to a "combinatorial" version of the repeated newsvendor problem.


Google's comment ranking system will be a hit with the alt-right

Engadget

A recent, sprawling Wired feature outlined the results of its analysis on toxicity in online commenters across the United States. Unsurprisingly, it was like catnip for everyone who's ever heard the phrase "don't read the comments." According to The Great Tech Panic: Trolls Across America, Vermont has the most toxic online commenters, whereas Sharpsburg, Georgia "is the least toxic city in the US." The underlying API used to determine "toxicity" scores phrases like "I am a gay black woman" as 87 percent toxicity, and phrases like "I am a man" as the least toxic. The API, called Perspective, is made by Google's Alphabet within its Jigsaw incubator.


Tool checks whether websites have built-in prejudice

Daily Mail

From reports Amazon's same-day delivery is less available in black neighbourhoods to Microsoft's'racist' chatbots, signs of online prejudice are becoming increasingly common. Scientists now say they can spot racist and sexist software using a code that finds out if there is implicit bias in algorithms running on websites and apps. By changing specific variables - such as race, gender or other distinctive traits - the online code Themis claims to know if data is discriminating against specific people. Previous research suggests technology is generally becoming racist and sexist as it learns from humans - and as a result, hindering its ability to make balanced decisions. Themis is a freely available code that mimics the process of entering data - such as making a loan application - into a given website or app.


Princeton researchers discover why AI become racist and sexist

#artificialintelligence

Many AIs are trained to understand human language by learning from a massive corpus known as the Common Crawl. The Common Crawl is the result of a large-scale crawl of the Internet in 2014 that contains 840 billion tokens, or words. Princeton Center for Information Technology Policy researcher Aylin Caliskan and her colleagues wondered whether that corpus--created by millions of people typing away online--might contain biases that could be discovered by algorithm. To figure it out, they turned to an unusual source: the Implicit Association Test (IAT), which is used to measure often unconscious social attitudes. People taking the IAT are asked to put words into two categories.


Pew Research Center: Internet, Science and Tech on the Future of Free Speech

#artificialintelligence

The more hopeful among these respondents cited a series of changes they expect in the next decade that could improve the tone of online life. They believe: Technical and human solutions will arise as the online world splinters into segmented, controlled social zones with the help of artificial intelligence (AI). While many of these experts were unanimous in expressing a level of concern about online discourse today many did express an expectation for improvement. These respondents said it is likely the coming decade will see a widespread move to more-secure services, applications, and platforms, reputation systems and more-robust user-identification policies. They predict more online platforms will require clear identification of participants; some expect that online reputation systems will be widely used in the future. Some expect that online social forums will splinter into segmented spaces, some highly protected and monitored while others retain much of the free-for-all character of today's platforms. Many said they expect that due to advances in AI, "intelligent agents" or bots will begin to more thoroughly scour forums for toxic commentary in addition to helping users locate and contribute to civil discussions. Jim Hendler, professor of computer science at Rensselaer Polytechnic Institute, wrote, "Technologies will evolve/adapt to allow users more control and avoidance of trolling.


Princeton researchers discover why AI become racist and sexist

#artificialintelligence

Many AIs are trained to understand human language by learning from a massive corpus known as the Common Crawl. The Common Crawl is the result of a large-scale crawl of the Internet in 2014 that contains 840 billion tokens, or words. Princeton Center for Information Technology Policy researcher Aylin Caliskan and her colleagues wondered whether that corpus--created by millions of people typing away online--might contain biases that could be discovered by algorithm. To figure it out, they turned to an unusual source: the Implicit Association Test (IAT), which is used to measure often unconscious social attitudes. People taking the IAT are asked to put words into two categories.


Princeton researchers discover why AI become racist and sexist

#artificialintelligence

Many AIs are trained to understand human language by learning from a massive corpus known as the Common Crawl. The Common Crawl is the result of a large-scale crawl of the Internet in 2014 that contains 840 billion tokens, or words. Princeton Center for Information Technology Policy researcher Aylin Caliskan and her colleagues wondered whether that corpus--created by millions of people typing away online--might contain biases that could be discovered by algorithm. To figure it out, they turned to an unusual source: the Implicit Association Test (IAT), which is used to measure often unconscious social attitudes. People taking the IAT are asked to put words into two categories.


Artificial intelligence: How to avoid racist algorithms

BBC News

There is growing concern that many of the algorithms that make decisions about our lives - from what we see on the internet to how likely we are to become victims or instigators of crime - are trained on data sets that do not include a diverse range of people. The result can be that the decision-making becomes inherently biased, albeit accidentally. Try searching online for an image of "hands" or "babies" using any of the big search engines and you are likely to find largely white results. In 2015, graphic designer Johanna Burai created the World White Web project after searching for an image of human hands and finding exclusively white hands in the top image results on Google. Her website offers "alternative" hand pictures that can be used by content creators online to redress the balance and thus be picked up by the search engine.