As San Francisco's residents shelter in place, coyotes are emerging on the streets; A Harvard Working Paper gives best practices for business to use in preparation for novel risks; A training simulator aims to make autonomous safer before encountering real-world roads; An article suggests using wastewater sampling to identify drug usage trends; and a Steem post offers two web sites with learn-at-home resources for students
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.
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.
pixabay license: source.
- Coyotes are roaming San Francisco's empty streets as the city's shelter-in-place order keeps people in their homes - Coyotes were native to San Francisco until hunting and poisoning nearly wiped them out in the 1940s. They made a recovery and were sighted in the area again during 2002, and now that the city is on lockdown, the animals are taking advantage of the vacant streets, and people are getting photographs and sharing on the Internet. According to a tour guide in the Wolf Sanctuary of Pennsylvania, the species has a similar history here in our state. During a tour a few years ago, the tour guide told us that coyotes have now returned to every county in the state of Pennsylvania. Click through to see the photos.
- Novel Risks - This Working Paper from Harvard Business School suggests that businesses can minimize the consequences from unanticipated risks by setting up teams to deal with unexpected circumstances. The Abstract acknowledges that all organizations know about the need to deal with certain kinds of foreseeable risk, but suggests that rigid policies for anticipated risks can fail when a novel risk appears due to circumstances that haven't been seen or thought of before. In particular, the abstract notes that certain types of biases can cause novel risks to go unnoticed and unmitigated. It goes on to describe best practices that some organizations are using to identify and manage these unforeseen risks.
- System trains driverless cars in simulation before they hit the road - MIT researchers have invented a photorealistic world that has infinite steering possibilities and can be used as a simulator for the training of autonomous vehicles. Other researchers have attempted simulation-based training environments, but past efforts have not been able to translate simulation success into success in the real world, in large part because acidents are rare, so limited data is available for simulating dangerous situations. In contrast, the new system from MIT uses photos collected by human drivers to synthesize a limitless range of possible scenarios that can be used for training. In tests, cars trained on this simulator were able to safely navigate roads that they hadn't encountered before and also to regain the road after a near-crash situation. The work makes use of reinforcement learning, and it was done in collaboration with the Toyota Research Institute. It is described in Learning Robust Control Policies for End-to-End Autonomous Driving From Data-Driven Simulation. Previous coverage of reinforcement learning in this series includes: Curating the Internet: Science and technology digest for January 23, 2020, Curating the Internet: Science and technology digest for January 22, 2020, and Curating the Internet: Science and technology digest for February 8, 2020. -h/t Communications of the ACM: Artificial Intelligence
- Want to Know if a New Drug Crisis Is Emerging? - Subtitle: Check the wastewater. The article asserts that policy makers were blind-sided by the speed with which fentanyl replaced heroin on America's streets, but that if someone had been monitoring wastewater, we could have had a real-time warning of the shifting drug usage patterns. Now, it says policy makers are becoming concerned about a resurgence of methamphetamine use, but that existing testing and monitoring methods - like surveys - are too slow. According to the article, wastewater sampling works because our bodies metabolize each of the drugs differently, so taking samples of wastewater and performing analysis on them can give policymakers an idea of the aggregate levels of drug usage in a region. Another advantage that wastewater analysis has on existing methods of tracking is that it is fast and cheap. Some cities in Washington and Oregon have already made early use of this technique. The article doesn't really address any potential privacy concerns, and it points out that the technique is not feasible in places with high percentages of individual household septic systems. The article closes by promoting a hypothetical, saying:
Imagine if a city not yet swamped by fentanyl had adopted wastewater testing. A spike in detected metabolites could have given everyone early warning. Law enforcement could have been on the lookout for suppliers. Public health authorities might have optimized the distribution of naloxone. Social workers might have warned drug users about the risks of heroin laced with fentanyl. Many lives might have been saved.
- Steem @bengy: Two Handy Schooling Websites for Kids (Coding and Physics) - This post tells us about two web sites for elementary and high school students. The first is code.org web site which is a free resource for learning to program with courses that are rated from K-12. According to the site, courses are centered on solving a particular topic. After introducing the topic, students are presented with 10 problems to solve. The second web site is the Einstein-First Project, which aims to give students an earlier exposure to modern physics concepts like quantum physics and general relativity. Apparently, the idea is that kids spend so much time and effort on Newtonian physics when they are younger that it impedes their later progress in more "advanced" physics concepts. Instead, David Blair launched this project so that kids can develop an intuition for physics at the large and small scales. Unlike code.org, the Einstein-First Project is still in a trial phase. (A 10% beneficiary setting has been applied to this post for @bengy.)
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.
This post will also appear on my pre-Steem blog, chescosteve.blogspot.com.
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.
- Cited author(s) - bengy - 10%
- Experimental - penny4thoughts - 36%
- Posting and/or scheduling service (steempeak.com) - steempeak - 5%
- STEM curation - steemstem - 5%
- SteemWorld.Org Tools for Steemizens - steemchiller - 5%
Sharing a link does not imply endorsement or agreement, and I receive no incentives for sharing from any of the content creators.