Discussion Topic |
|
This thread has been locked |
healyje
Trad climber
Portland, Oregon
|
|
|
Jul 13, 2017 - 08:15am PT
|
Jonathan: Computers are not “creative.” They maximize some numeric function. Calling a computer creative is a form of anthropomorphism.
This. Algorithms, whether brute force or not, simply 'weigh' more 'solution sets' than a human can, evaluate the 'fitness' of potential solutions more reliably than a human can, and are more 'thorough' than a human being so they will sometimes come upon things in a given situation that a human typically would not recognize or dismiss giving the appearance of novelty or of being 'creative'. It's not - it's just a chip / machine running code. Are they 'smarter' than a blender or a lawnmower? I suppose in the same way a calculator is 'smarter' than five year old, but it's the wrong way to look at such things in general.
Part of what's going on that lends the appearance of 'smartness' is sensors and automation. In the past we'd manually go look at a dial, feed the data into a computer, run a program, look at the results, and then go set a switch or valve to a specific setting. Now we automate such things so the computer software can continuously monitor a set of sensor parameters and do the switch / valve setting itself without a human in the loop. What actions we automate - i.e. what 'tasks' a computer can take on without human oversight - based on what parameters of a given reliability, with what checks, permissions and safeguards is the issue at hand and what exhibits the potential for 'harm'.
Another danger comes when we set a computer to a task without clearly understanding the entire domain and so fail to anticipate all the conditions that can occur in that space. As a result, computers need to have appropriate failsafes and failsafe protocols for such circumstances. In general there are inherent risks with letting 'smart' computers attempt to 'learn' in novel environments where we think we understand the domain better than we actually do. Self-driving cars being an prime example of one where it's hard to both acquire and evaluate sensor data and to anticipate all the environmental novelties and potential stupid human tricks of other real drivers on the road. Deploying them in mixed environments of both human and computer drivers will not be particularly pretty for the first few years. Also, if Toyota, GM, Fiat, Uber, etc. each deploy self-driving cars with different software to do the same task - driving - the results could similarly be quite unpleasant for a period.
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Jul 13, 2017 - 08:31am PT
|
I generally use single quote marks when quoting sections that have quotes embedded, that way you don't have an ambiguity about the beginning and ending of quotes.
The stylistic choice that has the punctuation embedded in a quote is just that, though there is a modern orthodoxy regarding the placement of punctuation in quotes. I generally try to adhere to the style, but issues of "use and mention" crop up and the logic of altering the "mention" by placing punctuation inside of it seems to be at least a stressful application.
Maybe we should get Chris to give you a red font so you can correct passages of incorrect punctuation usage.
|
|
healyje
Trad climber
Portland, Oregon
|
|
|
Jul 13, 2017 - 08:35am PT
|
I use single quotes for highlighting or to indicate a word may carry a concept or meaning different than typically conferred in the use of the word. I use double quotes for an actual quote - they are quite different uses and why we have each.
If your knickers get in a knot over such matters then online forums are going be a source of continuous agitation you might be better off avoiding. Being a grammar nanny or cop to this crew is akin to pointing out how they can correct small but inappropriate 'mistakes' in their climbing. Good luck with that.
|
|
jgill
Boulder climber
The high prairie of southern Colorado
|
|
|
Jul 13, 2017 - 02:47pm PT
|
"The above examples also show that the American style places commas and periods inside the quotation marks, even if they are not in the original material. British style (more sensibly) places unquoted periods and commas outside the quotation marks. "
I prefer the British style. But I still use the double quotation marks in "general". If I write this "sentence." I get the uneasy feeling I should have written "If I write this sentence."
But it's good to know you are a literary tech, and therefore are a white, middle age, paunchy man, Sycorax!
|
|
eeyonkee
Trad climber
Golden, CO
|
|
|
Jul 13, 2017 - 05:10pm PT
|
I like to wing it with quotes, like every thing else.
|
|
MH2
Boulder climber
Andy Cairns
|
|
|
Jul 14, 2017 - 07:31am PT
|
The self-driving car might need to confront a few of the same problems that the nervous systems of organisms from long ago, say 500 million years, were faced with: how to recognize objects around them and react appropriately .
For example, an old person has stepped off a sidewalk into a crosswalk. The car is approaching. The person ahead waves a cane. Are they telling the car to proceed or saying, "Hey! Can't you see I'm crossing the road?"
If humans create a car that can drive itself safely on our streets with us using the same streets, it would be a big step for machine "intelligence."
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Jul 14, 2017 - 08:55am PT
|
here's to winging it!
A raven's memories are for the future
Summary
The human brain stores memories of past events to guide decision-making about current and future events. Researchers long assumed that animals do not use memories in this way but rather exist in a constant stream of present needs, unable to plan for the future (1). Studies on nonhuman primates and corvids challenge this view and show that some species can plan for the future at least as well as 4-year old children (2, 3). These results suggest that planning for the future is not uniquely human and evolved independently in distantly related species to address common problems (4). On page 202 of this issue, Kabadayi and Osvath (5) show that ravens anticipate the nature, time, and location of a future event based on previous experiences. The ravens' behavior is not merely prospective, anticipating future states (6); rather, they flexibly apply future planning in behaviors not typically seen in the wild.
[Click to View YouTube Video]
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Jul 14, 2017 - 09:41am PT
|
|
|
Bob D'A
Trad climber
Taos, NM
|
|
|
Jul 14, 2017 - 10:14am PT
|
"Pinyon Jays have an expandable neck pouch that can carry over 50 seeds at a time. A flock will congregate and gather pinyon seeds. When enough birds have full throats, a verbal signal is given and the birds fly en masse to an open area, where they will walk shoulder to shoulder and poke the seeds into the ground much like farmers sowing crops. This cache will be used for winter survival. Researchers have proved that the jays can remember up to 95% of their cache locations, which leaves 5% of the cached seeds with the potential to germinate and grow new trees.
Pinyon Jays are monogamous; scientists’ efforts to get them to “cheat” on their mates in controlled experiments were unsuccessful. In the event of death of a mate, the survivor will often seek a new mate in other flocks or will help at the nests of his parents rather than raise a brood of his own."
|
|
jgill
Boulder climber
The high prairie of southern Colorado
|
|
|
Jul 14, 2017 - 12:26pm PT
|
Nice article you linked, Tim. I know very little about your area, but I learned of the existence of the American Institute of Mathematics by reading this. It was created about the time I retired.
|
|
Ward Trotter
Trad climber
|
|
|
Jul 14, 2017 - 12:27pm PT
|
I have been observing American Crows (Corvus brachyrhynchos) in the local environment for several years and have witnessed some interesting things and some enigmatic ones.
In the early 2000s their populations were devastated by the West Nile Virus:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2892874/
By the mid-2000s their numbers had declined precipitously relative to previous decades. Observant people growing up in Southern California must have noticed those strange mornings when the cawing of crows were rarely heard. On the cool overcast mornings you could easily hear the impressionistic staccato droning of band-tailed pigeons and the moaning legatos of the mourning doves-- but the raucous cawing of the crows had grown few and far between.
I'll never forget the day I spotted a young crow-- the first youngster I'd spotted in two or more years. Despite the fact this crow was ravenously ( no pun) attacking a discarded burrito he/she nevertheless appeared mighty lonely -- with not a fellow anywhere in sight. It was a sad thing to behold; but the consolation being that this smallish crow was young and apparently healthy and might represent a comeback of sorts. So I left it at that.
It was during this time I noticed two full grown crows nearby. Actually I first heard them from inside the house and their cawing was so loud and raucous that I rushed outside to check it out. They were engaged in what I have since come to think of as the " mockingbird vs crow" battles.
As you might be aware ,mockingbirds are extremely protective of their territories and therefore not only protect against other male mockingbirds but against squirrels, crows, and they'll even harry a human from time to time. They will dive bomb, sweeping ever lower on each pass, until the trespasser flees beyond a certain territorial point the mocker feels is acceptable.
Not long ago I witnessed a squirrel running along a telephone wire while being harried by a mocker . As soon as the squirrel made a left turn on intersecting wires that mocker backed off and another mocker in the adjoining territory took up where that one left off and harassed the poor squirrel until presumably yet another bird took up the challenge in this tag team harassment. The entire time my dog was also barking at the squirrel. This was a squirrel with a tough commute.
Gotta go.
I'll be back in due course to finish the " mockingbird vs crow" story.
Don't go away.
|
|
healyje
Trad climber
Portland, Oregon
|
|
|
Jul 14, 2017 - 01:31pm PT
|
And so you're going to provide some insightful pith of your own? Not sure if my waiting for some would count as 'bated' or 'baited'...
|
|
jgill
Boulder climber
The high prairie of southern Colorado
|
|
|
Jul 14, 2017 - 01:50pm PT
|
Zzzzzzzzzzzzz, antiseptically-written sciences passages
As opposed to the colorful but empty flow of literature, including philosophy. How many skillfully wrought passages do you think the Wizard has posted on no-thingness?
(What about 'Will' on TV? Marlowe has saved his ass!)
|
|
Bob D'A
Trad climber
Taos, NM
|
|
|
Jul 14, 2017 - 01:52pm PT
|
Zzzzzzzzzzzzz, antiseptically-written sciences passages
Like this one?
“Humanity is a biological species, living in a biological environment, because like all species, we are exquisitely adapted in everything: from our behavior, to our genetics, to our physiology, to that particular environment in which we live. The earth is our home. Unless we preserve the rest of life, as a sacred duty, we will be endangering ourselves by destroying the home in which we evolved, and on which we completely depend.”
― Edward O. Wilson
|
|
Largo
Sport climber
The Big Wide Open Face
|
|
|
Topic Author's Reply - Jul 14, 2017 - 08:21pm PT
|
If humans create a car that can drive itself safely on our streets with us using the same streets, it would be a big step for machine "intelligence."
Crickets from Camp Woo.
-
Dingus, there's hardly a reason to pluck every piece of low-hanging fruit on this thread. But see how the car does without you punching in a destination. The driverless car can get there because it's knows a tangible end point, and if it has access to any number of outfits that post live traffic conditions, it can adjust accordingly if it has all the maps (data) on board. But the whole thing hinges on having a fixed end point, a destination.
Now consider a creative task where the end point or destination or final product has to be sorted out as you go, with no existing maps, no distance known, no idea how the task will flesh itself out into the product. That's when an observer (awareness) and intelligence, comes into play.
There's just too much choss mixed in here with logically reasoned out evaluations from people looking to learn something new, or to broaden their understanding through meaningful discussion. I suppose I should expect nothing more from a mindset that writes about "the empty flow of literature, including philosophy." Empty of what? You got it, amigos - empty of calculations.
Shakespear, Austin, Blake, Chaucer, Dickens, Donne, Eliot, Milton, Coleridge - all empty. Not to mention Aristotle, Plato, Socrates, Kant, Jung, Hobbs, Whitehead - also, nothing but empty.
Mercy...
|
|
jgill
Boulder climber
The high prairie of southern Colorado
|
|
|
Jul 14, 2017 - 08:57pm PT
|
(Shakespeare, not Shakespear. The road to no-thingness begins with the loss of a single letter.)
Funny. I thought Paul or sycorax might want to bat my pitch back. Guess I've lost my touch. Oh well, back to 'antiseptically-written sciences passages'.
|
|
Ed Hartouni
Trad climber
Livermore, CA
|
|
|
Jul 14, 2017 - 08:59pm PT
|
Now consider a creative task where the end point or destination or final product has to be sorted out as you go, with no existing maps, no distance known, no idea how the task will flesh itself out into the product. That's when an observer (awareness) and intelligence, comes into play.
oh, like Mars Curiosity
NASA'S Mars Curiosity Debuts Autonomous Navigation
|
|
jgill
Boulder climber
The high prairie of southern Colorado
|
|
|
Jul 14, 2017 - 09:36pm PT
|
Yesterday, upon the stair,
I met a man who wasn’t there . . .
His software failed, his wiring bare,
One moment here, the next thin air.
|
|
|
SuperTopo on the Web
|
Let us know!