Awesome, not awesome.
#Awesome
“A group of researchers have now used [unsupervised machine learning] to munch through 3.3 million scientific abstracts published between 1922 and 2018 in journals that would likely contain materials science research. The resulting word relationships captured fundamental knowledge within the field, including the structure of the periodic table and the way chemicals’ structures relate to their properties. Because of the technique’s ability to compute analogies, it also found a number of chemical compounds that demonstrate properties similar to those of thermoelectric materials but have not been studied as such before. The researchers believe this could be a new way to mine existing scientific literature for previously unconsidered correlations and accelerate the advancement of research in a field.” — Karen Hao, AI Reporter Learn More from MIT Technology Review >
#Not Awesome
“In the case of the pedophile scandal, YouTube’s AI was actively recommending suggestive videos of children to users who were most likely to engage with those videos. The stronger the AI becomes — that is, the more data it has — the more efficient it will become at recommending specific user-targeted content. Here’s where it gets dangerous: As the AI improves, it will be able to more precisely predict who is interested in this content; thus, it’s also less likely to recommend such content to those who aren’t. At that stage, problems with the algorithm become exponentially harder to notice, as content is unlikely to be flagged or reported. In the case of the pedophilia recommendation chain, YouTube should be grateful to the user who found and exposed it. Without him, the cycle could have continued for years.” — Guiallume Chaslot, Researcher Learn More from WIRED >
What we’re reading.
1/ Companies might be using an image of your face to train their facial recognition software. Learn More from The New York Times >
2/ The US is renewing its push to create global standards for artificial intelligence regulation in its favor, not China’s. Learn More from Axios >
3/ An automated poker-playing algorithm can now beat the world’s best players in a multiplayer of no-limi Texas Hold ’Em poker — a major breakthrough for AI technology since the game is based on hidden information. Learn More from The New York Times >
4/ One of the world’s foremost leaders in AI research believes an “underrated deep-learning subcategory known as unsupervised learning” will lead to the next revolution in artificial intelligence. Learn More from MIT Technology Review >
5/ Based on a user’s search history, Google can predict if a person is a suicide risk and show them the National Suicide Prevention Lifeline. Learn More from The Atlantic >
6/ If Jon Steinbeck were still alive today, he might say that artificial intelligence, a “system built on a pattern [will] try to destroy the free mind, for this is the one thing which can by inspection destroy such a system.” Learn More from recode >
7/ Amazon is making a massive investment in re-training its employees who are likely to be automated out of their jobs by machine learning. Learn More from The Wall Street Journal >
What we’re building.
Come build the next-generation productivity platform with us at Journal! We’re hiring frontend, fullstack, infrastructure, and ML engineers. See the opportunities >
Links from the community.
“Data Science at The New York Times” submitted by Samiur Rahman (@samiur1204). Learn More from the Domino >
“Runway ML puts AI tools in the hands of creators everywhere” submitted by Avi Eisenberger (@aeisenberger). Learn More from The Verge >
First time reading Machine Learnings? Subscribe to get the next edition >
?First time reading Machine Learnings? Sign up to get an early version of the newsletter next Sunday evening. Get the newsletter >
Using your face to train AI was originally published in Machine Learnings on Medium, where people are continuing the conversation by highlighting and responding to this story.
Leave a Reply