[Popular STEM] Curating the Internet: Science and technology digest for October 3, 2020

2개월 전

IEEE Spectrum's weekly selection of awesome robot videos; An Indiana factory plans to reduce plastic pollution by converting it into eco-friendly fuel at commercial scale; AI helps artist to produce realistic, lifelike images of 54 Roman emperors; Algorithm spots COVID-19 infections in images of patients' eyes; and Research finds gender-preference differences that are independent of socio-environmental factors - suggests that women prefer working with people, men prefer working with things


Administrivia

Experimental: Please leave a relevant and substantive reply to this post in order to be considered for a gratuity from @penny4thoughts after the post pays out.

Fresh and Informative STEM Content: Welcome to my little corner of the blockchain

Straight from my RSS feed
Whatever gets my attention

Links and micro-summaries from my 1000+ daily headlines. I filter them so you don't have to.

First posted on the Steem blockchain: SteemIt.

image.png

pixabay license: source.

  1. Video Friday: An In-Depth Look at Mesmer Humanoid Robot - IEEE Spectrum's weekly selection of awesome robot videos for this week contains videos showing: a lifelike humanoid robot from Engineered Arts called, Mesmer; an autonomous serving robot from Bear Robotics; A video from security researchers at Brown who scanned the Internet and took control of some of the devices they found. They even found two of their own devices, and - with permission - took a compromised device on a drive from across the country; “(possibly) world’s first full-autonomous clear-up-the-table robot.” from Smile Robotics; plus more...

    Personally, I think that Mesmer's still stuck in the Uncanny Valley, but what do you think? Here he is:



  2. A factory in Indiana is turning plastic waste into eco-friendly fuel — and it's trying to revolutionize how we handle pollution - An Indiana factory that's operated by Brightmark is expected to shift into full time operation in 2021. The firm says it will convert plastic waste into wax and eco-friendly fuels at a commercial scale. The full process is a trade secret, but it begins with breaking up bales of plastic waste and then shredding them into pellets. At the end of the line, ultra-low sulfur diesel fuel emerges. Plant manager Jason Sasse is quoted, describing the plant employees as, "basically refinery operators." The company's web site also says that they are able to convert organic waste into renewable natural gas. Critics argue that plants like this have environmental costs of their own, but the firm responds that the net result is an improvement over the status quo, and that if something isn't done, we're currently on track to have more plastic than fish in the oceans by 2050. Of course, this isn't the first time something like this has been tried. Similar plants are already operating in Europe, but market conditions have prevented them from establishing a foothold in the US. Time will tell if it turns out better for Brighmark than it did for Changing World Technologies.

  3. AI 'resurrects' 54 Roman emperors, in stunningly lifelike images - Artist and cinematographer, Daniel Voshart used ancient sculptures, other artifacts, and machine learning to create lifelike images of 54 Roman Emperors, including Augustus, Nero, Hadrian, Caligula, and others. The machine learning depended on a neural network called, Artbreeder. As a first pass, Artbreeder took its input from about 800 busts to create realistic hair, skin and other facial shapes and features. After that, the models were refined using photoshop and details obtained from coins, artworks, and written descriptions from the time periods. It took Voshart about 2 months to track down the reference material, and each of the images took about 16 hours of processing. The full gallery is available on Voshart's blog. -h/t Communications of the ACM

  4. Algorithm Spots COVID-19 Cases from Eye Images: Preprint - In a September 10 preprint, scientists described an algorithm that was able to scan several hundred photos of people with and without COVID and to determine whether they were ill with more than 90% accuracy. Yanwei Fu is quoted as saying, "Our model is quite fast.", and "In less than a second it can check results." In comparison, current methods for identifying COVID involve expensive medical imaging or time consuming lab tests. Succinctly, the results are described like this:
    Of 24 people with confirmed coronavirus infections, the tool correctly diagnosed 23, Fu tells The Scientist. And the algorithm accurately identified 30 out of 30 uninfected individuals.
    However, the article also quotes opthalmologist, Daniel Ting who cautions that the sample size is too small to draw any conclusions, and the experiment would need to be repeated on at least 10 times as many patients. Another opthalmologist, Alastair Denniston suggests that an algorithm like this would have difficulty distinguishing between COVID-19 and other infectious forms of conjunctivitis, saying, "As an ophthalmologist it would be very surprising if there is a distinct COVID viral conjunctivitis pattern as opposed to other similar forms of viral conjunctivitis". -h/t Communications of the ACM Meanwhile, the Helsinki Airport is using dogs for fast COVID screening.

  5. Things versus People: Gender Differences in Vocational Interests and in Occupational Preferences - In a June study, the IZA Institute of Labor Economics found that women tend to prefer working with people whereas men prefer working with things, and that this variable is - in fact - the most important predictor of gender segregation in the workplace. The work also replicated this finding in the career aspirations for students in 8th and 9th grade, and lastly it found that this difference is "is largely independent of individual, parental, and regional controls." -h/t Daniel Lemire


In order to help bring Steem's content to a new audience, if you think this post was informative, please consider sharing it through your other social media accounts.


And to help make Steem the go-to place for timely information on diverse topics, I invite you to discuss any of these links in the comments and/or your own response post.

Beneficiaries


About this series


Sharing a link does not imply endorsement or agreement, and I receive no incentives for sharing from any of the content creators.

Follow on steem: remlaps-lite, remlaps


Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
STEEMKR.COM IS SPONSORED BY
ADVERTISEMENT