Slashdot Mirror
An Algorithm That Can Predict Human Behavior Better Than Humans (mit.edu)
Quartz describes an MIT study with the surprising conclusion that at least in some circumstances, an algorithm can not only sift numbers faster than humans (after all, that's what computers are best at), but also discern relevant factors within a complex data set more accurately and more quickly than can teams of humans. In a competition involving 905 human teams, a system called the Data Science Machine, designed by MIT master's student Max Kanter and his advisor, Kalyan Veeramachaneni, beat most of the humans for accuracy and speed in three tests of predictive power, including one about "whether a student would drop out during the next ten days, based on student interactions with resources on an online course."
Teams might have looked at how late students turned in their problem sets, or whether they spent any time looking at lecture notes. But instead, MIT News reports, the two most important indicators turned out to be how far ahead of a deadline the student began working on their problem set, and how much time the student spent on the course website. ...
The Data Science Machine performed well in this competition. It was also successful in two other competitions, one in which participants had to predict whether a crowd-funded project would be considered “exciting” and another if a customer would become a repeat buyer.
He wants us to die. It doesn't take an algorithm to understand that.
It might be interesting to try it at predicting future violent behavior of individuals.
Those who study such behavior have come up with the aphorism "The only effective predictor of future violent behavior is past violent behavior." Let's see if the
Shrinks who try to make predictions about individuals come out WORSE than chance - which implies that there may be SOME prediction possible - but the current paradigms have it backward.
This, by the way, is ONE of the reasons the pro-gun crowd pooh-poohs mental health tests for gun ownership or purchase. Another is the observation that people with mental illnesses are, on the average, far LESS likely to cause harm to others than the average of the population. (They may harm THEMSELVES, but suicide rates don't change if guns aren't available: Instead the suicidal switch to less effective and usually more painful means, averaging more tries before they succeed.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
(The Lenovo Trackpad ate my homework again. B-b )
Those who study such behavior have come up with the aphorism "The only effective predictor of future violent behavior is past violent behavior." Let's see if the program can identify any other indicators.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Just what we need right? More tracking of people and their (possible) habits. That surely won't be abused, used to pre-maturely judge / fire / expel / etc someone based on their social media pages (or lack thereof)? It won't be used to serve us ever more intrusive ads by monetizing every last detail of our lives and never giving us a penny? It could never be used to create a pre-crime system ala Person of Interest's "the machine"?
For everything these analytics programs do, I don't see much benefit from them. I DO see a fuckton of abuse. Is there any benefit to enhancing these beyond what they already are capable of beyond more privacy intrusion?
For those who complain about the heavy-handed privacy angle, you have to admit that's where this tech will be deployed and used the most. After all what's better than real world data created by making everyone an uncompensated and unaware test subject?
"MIT News reports, the two most important indicators turned out to be how far ahead of a deadline the student began working on their problem set, and how much time the student spent on the course website. "
IOW, the computer program used a more applicable approach/pertinent data and had humans had used the same algorithm/data, they would have reached the same or better results.
Where is the surprise? Oh wait. That's the second factor (time on course website). Which is kind of funny as it is such a weak metric, because people might use it offline or print it and don't necessarily need to be online. On the other hand it isn't really surprising that if you don't show interest you'll drop out.
Do we really have to read every MIT paper where they have applied an already existing theory?
The headline of the Quartz article and the Slashdot summary, "An algorithm can predict human behavior better than humans", is, not surprisingly, hugely overblown.
What these researchers actually did was develop a system for automatically taking a massive data set with a huge number of variables, identifying the subset of variables or new combinations of variables that are most likely to be useful for predicting a particular response, and then formulating a predictive model. (This is an extremely simplified summary.) That is really cool, but to present it as some sort of general "algorithm for predicting human behavior" is silly. It's no more an algorithm for predicting human behavior than are automated statistical methods for building a predictive model from a massive dataset.
but can it captcha?
(crickets)
Good! Maybe AT&T will finally use the system to realize we are sick and tired of their telemarketing calls and that we hate their slimy guts for their billing gimmicks.
Right now they seem clueless, as if a bunch of PHB's in a room stick a wet finger in.....the air to make a GUESS that we like being annoyed and that annoying us somehow makes them more profitable.
Robots couldn't do worse than the current marketing idiots there. Hell, give chimps a try also.
Table-ized A.I.
First, most humans are not that smart, but do not know that. It is called the "Dunning-Kruger" effect and it is well-established. Apparently at least the OP is unaware of it, possibly making him a subject of the Dunning-Kruger effect.
Second, a majority of humans do not use what smarts they have effectively, but rather do decide "emotionally" when it comes to important decisions or understanding important situations. That obviously works rather badly, just look at what politicians get voted into office, or what life-choices people make. The problem here is that the whole "emotional decision" apparatus is a rather primitive left-over from caveman-times that cannot handle even situations of moderate complexity well. The second problem is that most humans never find that out, as the skill for self-reflection is also rather scarce and hence cannot actively compensate.
So give an arbitrary group of people an analysis or decision problem that somehow "touches" them (like asking students to predict whether other students fail at being students), and suddenly most of them turn into morons (or rather do not stop being morons in many cases), and even a simplistic statistic predictor does a lot better than they do.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
You are all cows. Cows say moo. MOOOOOO! MOOOOO! Moo cows MOOOOO! Moo say the cows. YOU HUMAN COWS!!
A different algorithm developed by yep University of Illinois researchers, also outdo humans in the sense that bots running that algorithm can predict human's next move ...
http://www.defenseone.com/tech...
I am not the chicken little type, but with bots which know human better than human know themselves, this could be the start of some very troubling time for the genus Homo sapien Sapien
Muchas Gracias, Señor Edward Snowden !
There is one: Membership in the biological group "Homo Sapiens". Anything more accurate is just trying to stick labels on people to make them "different" and push them out of society. Some people like doing that to elevate their pathetic selves by stomping on people they perceive to be even more pathetic.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
It is impossible to behave as humans better than what humans can do. As far as predicting is basically simulating the given behaviour, there is nothing which can predict human behaviour better than what humans can do.
Translation of the previous paragraph to a more-appealing language (c-biased pseudo-code):
Function main()
{
behaviour = doBehaviour(true);
predAlgorithm = doBehaviour(false);
predHuman = doBehaviour(true);
countAlgorithm = 0;
countHuman = 0;
while (all elements in behavior/predAlgorithm/predHuman)
{
if(behavior_item == predAlgorithm_item) countAlgorithm ++;
else if(behavior_item == predHuman_item) countHuman ++;
}
If(countAlgorithm > countHuman) print “Review the whole algorithm because this result doesn’t make any sense”;
}
function doBehaviour(isHuman)
{
if(isHuman) return behave();
else return guessHowHumansBehave();
}
Custom Solvers 2.0 = Alvaro Carballo Garcia = varocarbas.
It would be interesting to see if the algorithm can come up with less obvious connections.
Say for example that there is a connection between date of birth and violent behavior, this could indicate that society treats people born in one part of the year differently, but it isn't something that one can easily guess.
That is sort of the entire point of this. Any suggestions in the comment field or in public debate will be of things that humans can easily predict. What they are looking for is connections that weren't obvious but should be.
In the past, what came out of these efforts war universally horrible. Like people of a certain descendant being regarded as of low quality and better be disposed of (3rd Reich), people of a certain skin color being regarded as dumb or violent (USA, many others), etc.
On the other hand, nothing useful was ever found. The only ethical thing is to stay away from the whole approach. It only feeds racism and similar mind-sets.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Asimov had it right...
I don't think Hari Seldon is too worried for his job.
To all the anti-gun nuts -
While I understand that your life is so boring that you just have to go around making up all kinds of silly lies ... but please, if you need to lie, please tell lies that are believable !
Do you know that South Korea has a suicide rate which is more than double of that of the United States?
Do you know Japan has a suicide rate of more than 60% higher of that of the United States?
Unlike the Americans, average South Koreans and Japanese don't own guns
http://www.washingtonpost.com/...
Here is a list of countries with their respective suicide rate
Muchas Gracias, Señor Edward Snowden !
Most humans has no creative capability, and value conformity over originality, even if conformity is not imposed.
Those of us who is not like them actually succeed and LEAD.
based on student interactions with resources on an online course.
Clearly the human brain is good at interpreting a different kind of behaviour (instantaneously and visually received), not temporal behaviour of interaction with a machine presented as a spreadsheet. Given the type of information and unintuitive presentation for the human brain, the question is: did the humans have as long with the training data as the ANN? I'm guessing not, but i suppose the competition does at least prove it's utility over an untrained human rather than to purely serve as publicity.
This is the sort of task we should expect artificial NN to be good at, because currently they are many orders of magnitude slower and smaller than biological NN, so their strength will be in artificial esoteric tasks that the human brain has not evolved (optimised) for.
We had long known that given a feature set, computers can automatically fit a pretty good model. Now, can they automaticall extract this feature set and data without humans?
Online gambling sites already know this and prepared to put their money up front to hook loosers.
They also quickly recognize when to black ban punters who seem to be professional real fast!
I suspect porn sites have you covered too, although I am not experienced. Are either different from amazon after you browse for baby items?
Prey on gullible and vulnerable has not changed at all, and they need less people to figure out who - a refinement n Nigerian scammers who rely on volume to lure suckers in.
Your wife will betray you tomorrow at 2:00 clock while you're at work.
connection between date of birth and violent behavior
Let's divide the duration of the year in say, 12 periods, and give it a scientific sounding name, for instance Astrology...
Non-Linux Penguins ?
But only within the confines of the experiment. Alas, human beings and life on earth are capricious and don't always act in alignment with spontaneous words after reflection has informed a course of action a bit more. Intelligence isn't consciousness, and data collation isn't critical thinking. It would be a slippery slope to presume otherwise and abdicate responsibility for our lives snd thinking. Look no further than corporations flip-flopping around like scared, dying fish in the early days of the web, they got their asses kicked and it's because they presumed infallability in their approaches. This is nonsense. And i won't register here until there is a way for users to delete accounts. ;)
Can drive nails into wood better than humans. I, for one, welcome our new hammer overlords.
look for things that correlate. Aren't there statistical functions that will give you that.
There are many cases where a baby rolling dice will outperform the average human in predicting decisions.
There are correlations with the time of year someone was born and later actions or reactions. Frequently this is attributed to the prevalence of the flu at certain times of year, or how old the child was relative to the other kids in their class. The evidence for causation, however, is usually as weak as that of astrology, even though the explanations do make currently acceptable sense.
Don't be too quick to deny patterns noticed by our ancestors, even if you doubt the reasons that they attributed to them. (OTOH, astrology was mainly about when to plant crops. Personal astrology never seems to have been very accurate, even if the Babylonians did use it to tell the king when to change is clothes.)
I think we've pushed this "anyone can grow up to be president" thing too far.
In the past, what came out of these efforts war universally horrible.
What I'm expecting, or at least hoping for, is for this impartial algorithm to fail to identify any strong predictors (other than previous history of initiated violence). Or at least not find anything far enough above statistical noise to be used as an excuse for government to deny a person the fundamental rights to self-defence from threat to life or limb - (either individual threats or a government disarming its own opposition before implementing tyranny or genocide.)
If it turns out that history of initiated violence is the ONLY substantial predictor it can find, that would argue for the only reason for abridging a person's RKBA would be "Ajudicated, by a court of competent jurisdiction, as being a danger to others, on the basis of a history of unjustified violent attacks against innocent parties." Even most RKBA advocates can get behind that. (Of those that don't, their reasoning would be things like: "We can't allow a potentially tyrannical government ANY excuse to, through malice or incompetence, disarm its potential righteous opposition." or "If this guy really IS a danger to others he should still be in jail or committed to a mental institution.)
If it DOES find some strong predictor, we'd then have some decent data on which to discuss what, if anything, should be done about "risky people". That's FAR better than the current laws and proposals, which amount to disarming people (and thus denying them a civil right) arbitrarily, on the basis of unsubstantiated - and so far proven incorrect - fads among the mental-health segment of the medical community and various political factions.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Dumb statistical correlations without real explanations are not science, and only politicians with an ideological bent can support them. Models and AI discriminators are no more than correlations. They are inductive, not instructive. I suggest you read about Russell's chicken. The chicken got fed every day and deductively formed the model of the farmers behavior, that the farmer was benign and loved chickens. The next day the farmer wrung the chicken's neck. So much for inductive reasoning. That chicken needed a better explanation for the farmer's behavior. So does that algorithm.
who's behind gun control. same tribe that has trillions in weapons. all the 'school shootings' are false flags so was 'boston marathon' bs. Trolls dumb down the situation. Many posts are also computer generated fakes. Know who owns you. .- thezog.info
holo fraud - https://archive.org/details/TheLeuchterReport
http://jewishcrimenetworkdid911.blogspot.com - that too.
holodomorinfo.com - see pages don't waste time on videos, sites even 'jew truther' sites run by them so you sit 'reading' or 'follow'.
Know who is behind the schemes. copy links, give to others.
also - have click show all comments button - and also slide bar over at top to see all posts. have to do Both to see all posts. notice what they hide. see hidden posts at top