New Articles (Page 382)
To stay up to date you can also follow on Mastodon.
π Moravec's Paradox
Moravec's paradox is the observation by artificial intelligence and robotics researchers that, contrary to traditional assumptions, reasoning (which is high-level in humans) requires very little computation, but sensorimotor skills (comparatively low-level in humans) require enormous computational resources. The principle was articulated by Hans Moravec, Rodney Brooks, Marvin Minsky and others in the 1980s. As Moravec writes, "it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility".
Similarly, Minsky emphasized that the most difficult human skills to reverse engineer are those that are unconscious. "In general, we're least aware of what our minds do best", he wrote, and added "we're more aware of simple processes that don't work well than of complex ones that work flawlessly".
Discussed on
- "Moravec's Paradox" | 2023-06-10 | 13 Upvotes 4 Comments
- "Moravec's Paradox" | 2019-08-15 | 155 Upvotes 87 Comments
- "Moravec's paradox" | 2018-04-21 | 30 Upvotes 6 Comments
- "Moravec's paradox" | 2016-02-06 | 30 Upvotes 4 Comments
- "Moravec's paradox" | 2012-12-14 | 188 Upvotes 43 Comments
π Tunnel of Eupalinos
The Tunnel of Eupalinos or Eupalinian aqueduct (Greek: ΞΟ Οαλίνιον ΟΟΟ Ξ³ΞΌΞ±, romanized:Β Efpalinion orygma) is a tunnel of 1,036Β m (3,399Β ft) length running through Mount Kastro in Samos, Greece, built in the 6th century BC to serve as an aqueduct. The tunnel is the second known tunnel in history which was excavated from both ends (Ancient Greek: αΌΞΌΟΞ―ΟΟΞΏΞΌΞΏΞ½, romanized:Β amphistomon, "having two openings"), and the first with a geometry-based approach in doing so. Today it is a popular tourist attraction.
Discussed on
- "Tunnel of Eupalinos" | 2023-05-06 | 195 Upvotes 32 Comments
π Q-Learning
Q-learning is a model-free reinforcement learning algorithm to learn a policy telling an agent what action to take under what circumstances. It does not require a model (hence the connotation "model-free") of the environment, and it can handle problems with stochastic transitions and rewards, without requiring adaptations.
For any finite Markov decision process (FMDP), Q-learning finds an optimal policy in the sense maximizing the expected value of the total reward over any and all successive steps, starting from the current state. Q-learning can identify an optimal action-selection policy for any given FMDP, given infinite exploration time and a partly-random policy. "Q" names the function that returns the reward used to provide the reinforcement and can be said to stand for the "quality" of an action taken in a given state.
Discussed on
- "Q-Learning" | 2019-08-13 | 88 Upvotes 33 Comments
π Michael Cicconetti
Michael A. Cicconetti (born 1951) is a retired Municipal Court judge who presided in Painesville, Lake County, Ohio, United States, dispensing a unique brand of what he calls creative justice. The judge often left the choice of penalty to the defendant, who was faced with spending time in jail or undergoing one of Cicconetti's unusual punishments. These often involved placing the defendant in a similar position to that of the defendant's victim at the time of the crime.
Cicconetti's first creative sentence, which involved a violation relating to a stopped school bus, occurred in the mid-1990s. Famously he offered 26-year-old Ohio housewife Michelle Murray the option (in return for a reduced prison sentence) of spending a night in the woods for abandoning 35 kittens in a forest in wintertime; he said: "You don't do that. You don't leave these poor little animals out and, yes, I wanted to set an example for her future conduct or anybody else who was contemplating doing such a thing". On other occasions he ordered noisy neighbors to spend a day of silence in the forest or listen to classical music instead of rock. In all cases the judge attempted to place a link between the perpetrated offense and its punishment.
Due in part to the popularity of his actions, he won the presidency of the American Judges Association. He attributes his unusual approach to his background. He is an Eagle Scout, earning the award in 1964, as a member of Scout Troop 64 in Painesville, Ohio. He was the oldest of nine siblings who had to work on ore boats throughout the Great Lakes as a deckhand and deckwatch to fund himself through college. After graduating from St. Leo University, he became Clerk of the Painesville Municipal Court while attending Cleveland State University Law School at night.
Many of the victims, but also defendants, claim that his unusual approach has helped them to cope with their problems and the judge is reportedly inundated with letters from his admirers. Furthermore, where the national recidivism (repeat offender) rate is over 75%, the rate in Judge Cicconetti's court was just 10%.
His philosophy is exemplified by the following two quotations:
When you engage people and praise them for their good behavior, not unlike children, it helps their self-esteem. My judicial philosophy is really not that much different from a parental philosophy. I have five children. You can paddle them or spank them but what do you gain? Most people want to be good but for little obstacles or habits. We have to change the habits and remove the obstacles. That's our job.
Sentences such as Cicconetti's are becoming more popular across the United States, and one judge has cited him specifically as being the influence for one of her own sentences.
In February 2019 Cicconetti announced that he planned to retire later in the year. He retired from being a judge on September 20, 2019.
Discussed on
- "Michael Cicconetti" | 2019-08-13 | 201 Upvotes 89 Comments
π Ad Creep
Ad creep is the "creep" of advertising into previously ad-free spaces.
The earliest verified appearance of the term "ad creep" is in a 1996 article "Creeping Commercials: Ads Worming Way Into TV Scripts" by Steve Johnson for the Chicago Tribune, however it may have been coined by a subscriber to Stay Free! magazine, according to another source.
While the virtues of advertising can be debated, ad-creep often especially refers to advertising which is invasive and coercive, such as ads in schools, doctor's offices and hospitals, restrooms, elevators, on ATMs, on garbage cans, on vehicles, on restaurant menus, and countless other items. In Steve Johnson's piece referenced above, he criticizes product placement and "creative advertising enhancements" as "one more manifestation of an environment in which the commercial assault is almost nonstop". Commercial Alert, a nonprofit organization founded by Public Citizen "to keep the commercial culture within its proper sphere, and to prevent it from exploiting children and subverting the higher values of family, community, environmental integrity and democracy" also characterizes "ad creep" as an assault, with ad companies fighting a "relentless battle to claim every waking moment, and what one executive called, with chilling candor, mind share". A 2017 Daily Express story in the UK suggests "the creeping incursion of adverts in Windows 10" has been an issue.
On the other hand, modern advertisers are compelled to react to changes in consumer habits. An article in The New York Times notes that "consumersβ viewing and reading habits are so scattershot now that many advertisers say the best way to reach time-pressed consumers is to try to catch their eye at literally every turn." And, the article suggests that ad agencies believe that as long as ads are entertaining, people may not mind the saturation. As people have turned from traditional media, advertisers have not only struggled to create brand awareness, but there is also a move to "microtarget people at precisely timed moments" as well, according to an article in Stay Free!.
Occasionally, the term "Ad Creep" has been used to describe a process of slowly infusing more ads into places where ads have been expected (television shows, for example) such as in a 2011 Advertising Age article describing the increase in both the time devoted to ads and the number of ad messages in the Super Bowl. This is not a standard use of the term, but it is related. A 2017 blog post by the chief global analyst of Kantar Millward Brown, a marketing firm, notes "that average ad loads on national television in the U.S. continued to creep upwards from 10.4 minutes per hour in December 2014, to 10.9 minutes in December 2016". Although the increase is less than 5%, he suggests "marketers should be concerned because the evidence suggests that more clutter is a bad thing for brands."
Discussed on
- "Ad Creep" | 2019-08-13 | 20 Upvotes 2 Comments
π List of selfie-related injuries and deaths
This is a list of serious injuries and deaths in which one or more subjects of a selfie were killed or injured, either before, during or after having taken a photo of themselves, with the accident at least in part attributed to the taking of the photo.
The United States Department of Transportation estimated that during 2014, the so-called "year of the selfie", 33,000 people were injured while driving and using a cell-phone in some fashion, which can include talking, listening, and "manual button/control actuation" including taking, uploading, downloading, editing, or opening of selfies. A 2015 survey by Erie Insurance Group found that 4% of all drivers admitted to taking selfies while driving.
The Washington Post reported in January 2016 that "about half" of at least 27 "selfie related" deaths in 2015 had occurred in India. No official datasets on the number of people who died taking selfies in India exists, but reports show from 2014 up to August 2016, there have been at least 54 deaths in India while taking selfies. The Indian Ministry of Tourism asked states to identify and barricade 'selfie danger' areas, its first national attempt to deal with the selfie deaths. Mumbai Police identified at least 16 danger zones after a man drowned attempting to save a selfie-taker. No-selfie zones were also established in certain areas of the Kumbh Mela because organizers feared bottlenecks caused by selfie-takers could spark stampedes.
A 2018 study of news reports showed that between October 2011 and November 2017, there were 259 selfie deaths in 137 incidents reported globally, with the highest occurrences in India, followed by Russia, United States, and Pakistan. The mean age was 23 years old, with male deaths outnumbering female about three to one.
Discussed on
- "List of selfie-related injuries and deaths" | 2019-08-12 | 50 Upvotes 42 Comments
- "List of selfie-related injuries and deaths" | 2016-06-05 | 25 Upvotes 2 Comments
π Creeping normality
Creeping normality (also called landscape amnesia) is a process by which a major change can be accepted as normal and acceptable if it happens slowly through small, often unnoticeable, increments of change. The change could otherwise be regarded as objectionable if it took place in a single step or short period.
American scientist, Jared Diamond, first coined the phrase creeping normality in his 2005 book, Collapse: How Societies Choose to Fail or Succeed. Prior to releasing his book, Diamond explored this theory while attempting to explain why, in the course of long-term environmental degradation, Easter Island natives would, seemingly irrationally, chop down the last tree:
"I suspect, though, that the disaster happened not with a bang but with a whimper. After all, there are those hundreds of abandoned statues to consider. The forest the islanders depended on for rollers and rope didn't simply disappear one dayβit vanished slowly, over decades."
Discussed on
- "Creeping normality" | 2019-08-10 | 218 Upvotes 73 Comments
π A*
A* (pronounced "A-star") is a graph traversal and path search algorithm, which is often used in computer science due to its completeness, optimality, and optimal efficiency. One major practical drawback is its space complexity, as it stores all generated nodes in memory. Thus, in practical travel-routing systems, it is generally outperformed by algorithms which can pre-process the graph to attain better performance, as well as memory-bounded approaches; however, A* is still the best solution in many cases.
Peter Hart, Nils Nilsson and Bertram Raphael of Stanford Research Institute (now SRI International) first published the algorithm in 1968. It can be seen as an extension of Edsger Dijkstra's 1959 algorithm. A* achieves better performance by using heuristics to guide its search.
Discussed on
- "A*" | 2019-08-10 | 50 Upvotes 6 Comments
π M44 (Cyanide Device)
The M44 cyanide device (also called a cyanide gun or cyanide trap) is used for the killing of coyotes, feral dogs, and foxes. It is made from four parts: a capsule holder wrapped with cloth or other soft material, a small plastic capsule containing 0.88 grams of sodium cyanide, a spring-powered ejector, and a 5-7 inch stake. To install the trap, the stake is first driven down into the ground, and then the capsule is put in the holder, screwed onto the cocked ejector, and secured to the stake. The wrapped capsule holder is smeared with scented bait to attract coyotes and make them bite and pull on it. (The use of a bite-and-pull action makes the trap less likely to be set off by non-canine wildlife.) When the trap is triggered, the spring propels a dose of sodium cyanide into the animals's mouth, and the sodium cyanide combines with water in the mouth to produce poisonous cyanide gas. In addition to the cyanide, the capsule contains Day-Glo fluorescent particle marker (orange in capsules used by the Wildlife Services, and yellow in capsules prepared for other users).
Discussed on
- "M44 (Cyanide Device)" | 2019-08-09 | 70 Upvotes 22 Comments
π 1957 German Grand Prix
The 1957 German Grand Prix was a Formula One motor race held on 4 August 1957 at NΓΌrburgring. It was race 6 of 8 in the 1957 World Championship of Drivers. The 22 lap race was won by Juan Manuel Fangio, and is often cited as one of the greatest victories in racing history. It was Fangio's fourth victory out of the seven races in the season contested by Formula 1 cars - excluding the Indianapolis 500, in which only US drivers competed, using USAC Championship cars. Furthermore, due to the number of points he had accumulated in the season (34 to Luigi Musso's 16), his victory at the NΓΌrburgring mathematically clinched Fangio's fifth World Championship title with two races to go. The race was also notable for being Fangio's 24th and last victory in F1; his career still stands as having the highest win percentage ever, with 46.15% of his 52 race entries being wins.
Discussed on
- "1957 German Grand Prix" | 2019-08-09 | 14 Upvotes 3 Comments