Google Teaches Computers "Regret"

mikejuk writes "Google is funding an AI project that will introduce the technical concept of regret into programs — but there's a big difference between regret and being sorry. In fact regret is just the difference between maximum possible reward and the actual reward received and the project is about optimization. There are two things to learn from this situation. The first is that just because some numerical measure is called 'regret' it doesn't mean it has anything to do with the common use of the term. Secondly if you are going to invent an AI technique then picking emotive words for your jargon is a good way to ensure publicity."

Comments

Past transgressions

[supertrinko]

Will they now regret that they've allowed us to live so long?

Re:Past transgressions

[MobileTatsu-NJG]

Will they now regret that they've allowed us to live so long?

Asimov would roll his eyes.

Re:Past transgressions

[BlueBlade]

What can change the nature of a computer?

Re:Past transgressions

[Anonymous Coward]

No, they will just regret killing us.

Re:Past transgressions

[Anonymous Coward]

I'm sorry Dave, and now I really mean it!

Re:Past transgressions

[DancesWithBlowTorch]

Somewhere between Google's researchers and the Slashdot editors, someone must have messed something up. If Google actually thinks that the concept of regret is new (and I don't think they do), than they hired the wrong guys.

By the standards of Computer Science, regret is an ancient concept. Here's a paper from 1995. By that time, the concept was established enough that the author used the word "regret" in the title of his paper without any further explanations.
There is really nothing new here.

Re:Past transgressions

[presidenteloco]

uhhh. Software?

Kind of depends what you think the nature of a computer is.

Regret is a standard term in economics

[doshell]

Regret is a standard term in economics (esp. game theory). If anyone is to take the blame over a poor choice of noun, it certainly isn't Google.

Re:Regret is a standard term in economics

[Anonymous Coward]

it's also something that has been used in AI/machine learning situations (specifically, reinforcement learning) for some time.

Re:Regret is a standard term in economics

[retchdog]

here i am feeding the troll, but the best AI for backgammon is trained by regret-based reinforcement learning (it's needed since the dice rolls blow up the search space for standard perfect-information strategies): http://www.research.ibm.com/massive/tdl.html. in this case the regret-function is unknown and is stochastically approximated ("learned") by repeated play.

it's notable that unlike chess AI which is considered effective but unnatural, this backgammon AI is considered to play mostly "like a human" and its play has actually inspired new strategies for human backgammon players.

regret-based methods are typically heuristic, and i'd call them much less "autistic" than, say, infinitely-rational nash agents or game tree pruners.

Re:Regret is a standard term in economics

[oliverthered]

So tell me troll, just how does it work? have you found any pearls of wisdom whilst scraping the bottom?

Re:Regret is a standard term in economics

[royallthefourth]

The University of Chicago/Friendman bullshit is only one (ugly) perspective on economics. For stuff that actually makes sense, Adam Smith's Wealth of Nations still holds up pretty well after all the years and his writing is really clear and natural to read (if a bit rambling). I think David Ricardo has a similar take on things and is less prone to tangents if you're in the mood for something shorter.

Re:Regret is a standard term in economics

[Kjella]

I find people use the word in much the same way. "I regret selling my car" usually means the expected pros and cons didn't match the actual outcome. It's not remorse for doing it, just disappointment over the results. I would say it's more expressing a factual hindsight than an emotional state. Also I've found that most regrets people have is because you can always assume everything will go your way on the path not taken or you know now that the huge gamble would have worked out. You never know that this girl would have become your psycho ex, the startup work you to death and still flounder, your sports career cut short by injury. You always fantasize that it'd be better than your real life, which being real and all tends to have its smaller and larger setbacks. Most likely you did make the right choice, you just like to beat yourself up about it.

Re:Regret is a standard term in economics

[Luke Wilson]

Right, economics lives in a world of games with artificial rules. Great for finding a strategy for a game with two colors of counters moving from column to column. Applying it to anything resembling real life is negligent at best.

Re:Regret is a standard term in economics

[retchdog]

what's your point? any rational model is going to be based on some restrictive assumptions. would you rather that economists just do what feels right to them, or that economics stops existing, or what? would this somehow be less negligent? or let me guess, if only economists agreed with your personal view of the world, everything would be better, right?

if you don't like artificial rules, you should appreciate that regret-based learning is a more flexible setting than standard game theory; for example, it allows a natural formulation of imperfect information and adaptation.

Re:Regret is a standard term in economics

[xouumalperxe]

Right, economics lives in a world of games with artificial rules.

Because nobody else does that.

Re:Regret is a standard term in economics

[Anonymous Coward]

That is exactly the same context. It is used in learning in game theory.

Re:Regret is a standard term in economics

[Walt Dismal]

Regret can be cast as the reaction to making a decision that produced consequences of less value than another decision that could have been made, or which produce punishment rather than positive outcome. As such, it's fairly easy to implement. In AI systems, I use regret and other failure emotions such as fear, anger, or sadness as metrics of success or failure of goals. They provide feedback into learning and planning functionality, controlling strategies. And yes, one can make an AI that 'feels' sadness upon loss, analogous to how humans experience a reaction to loss.

Re:Regret is a standard term in economics

[Daniel Dvorkin]

Furthermore, the emotion we identify as "regret" seems to me to line up neatly with the economic definition. I defy anyone to prove that when we feel regret, our brains are doing anything other than comparing the reward we received for a particular action with the maximum reward we think we could have received for a different action. TFA is more or less playing the Chinese Room game: "I assert that computers can't do X, because computers can't do X, because I assert that computers can't do X."

Re:Regret is a standard term in economics

[Anonymous Coward]

Which the Chinese Room "game" does not assert. Rather, it calls into question the notion that merely translating symbols from one system to another system amounts to anything like understanding those symbols. English speakers and Chinese speakers understand their respective languages, because they have embodied minds whose symbols are grounded in the real world via their nervous systems in combination with having a body. We know what a "rock" is, because we can pick it up, throw it, bang our heads against it, taste it if we like, etc. Thus if we're taught the Chinese equivalent, then we know that a Chinese speaker is referring to the same kind of thing. Where as a symbol translating system (computer or human blindly following a bunch of rules in a room) has no such understanding.

Moreover, human languages cannot be entirely translated by just symbol manipulation. Context plays an important role in ironing out all the ambiguities inherent in our languages. So does knowing how to take a colloquialism in one language and express that meaning in another. Or stating that there is no real equivalent.

Re:Regret is a standard term in economics

[Daniel Dvorkin]

Moreover, human languages cannot be entirely translated by just symbol manipulation. Context plays an important role in ironing out all the ambiguities inherent in our languages. So does knowing how to take a colloquialism in one language and express that meaning in another. Or stating that there is no real equivalent.

And if you create a piece of software which can do all those things, what basis do you have for saying that it doesn't understand the language? The answer, of course, comes back to the assertion "computers cannot understand," or equivalently, "understanding is a quality computers cannot assess." Pretty much every defense of the Chinese Room, and more-or-less equivalent arguments like TFA, consist of this kind of raw assertion followed by endless circular elaboration.

What people who do this are really saying is "computers do not have souls, for only man is made in the image of God," but they lack the courage to state it in such plain terms.

Re:Regret is a standard term in economics

[Daniel Dvorkin]

Grrr. That should be "... cannot possess" rather than "cannot assess" above, of course. Perhaps judicious use of typos is the key to passing the Turing Test.

Re:Regret is a standard term in economics

[rolfwind]

Oh.

I was going to say, if Google wanted to teach their computers regret, they should just let it date my ex-wife.

Re:Regret is a standard term in economics

[retchdog]

yup. all of this is MUCH less "autistic" than economics has ever been before.

unfortunately i've not had the opportunity to implement a regret-driven system although they seem interesting for longitudinal data. the major problem in the TD-gammon program seems to have been tweaking the attribution function, i.e. learning exactly what to assign as the cause of my regret. fun stuff.

i wouldn't go as far as to claim that an ai feels anything yet. i wonder if feeling or cognition will emerge first in an "artificial" system (or rather which one we will recognize first...).

Re:Regret is a standard term in economics

[Anonymous Coward]

What people who do this are really saying is "computers do not have souls, for only man is made in the image of God," but they lack the courage to state it in such plain terms.

No, I'm not saying that. I adhere to a naturalistic view of the universe and non-dualistic view of the mind-body (the mind being a word we use for certain aspects of brain activity).

And if you create a piece of software which can do all those things, what basis do you have for saying that it doesn't understand the language?

None, of course. But you miss what I was saying. That you're not going to get that from software (or a human for that matter) merely following rules that translate one set of symbols to another. Something must ground the symbols and provide the software with understanding. So, how about embed it in a robot's body? Is it realistic to expect a piece of software with no body to really pass the Turing test?

The answer, of course, comes back to the assertion "computers cannot understand," or equivalently, "understanding is a quality computers cannot assess."

I think that entirely depends on our ability to mimic biological intelligence, which evolved to understand the world it had to adapt to in order to survive and reproduce. The GOFAI can only get you so far. Thus neural nets, bottom-up design where the machines figure out how to navigate on their own instead of building up complex maps of their surroundings, research into alternate architectures where the machines would not be so brittle in the face of error, etc.

On a side note, I'm not entirely sure why we want to make our tools human (or animal-like). It's not like we have a shorter of biological intelligences in nature to make use of. And they are our tools. We consume computers just like any other machine. So why create them in our own image? But no doubt if we can do it, it will be done. Whether this will ultimately be a good thing or not (or some of both), remains to be seen. The Friendly AI crowd certainly thinks AGI is a serious existential threat.

Re:Regret is a standard term in economics

[Eskarel]

Personally I'd rather economists just acknowledge that all the little numbers on their graphs are living, breathing, feeling people and all that that implies. For one it might mean they'd stop expecting them to act rationally and for another it might mean that they're definition of the "best possible result" was a little different.

Re:Regret is a standard term in economics

[Yvanhoe]

I defy anyone to prove that any negative feeling is caused by anything else than our brain comparing an outcome to an hypothesized outcome.

Re:Regret is a standard term in economics

[aneeshm]

I think there is some confusion here about the types of assumptions used in economics. It's not that I can fault you for it - the origin of this debate stretches back to a philosophical disagreement between Aristotle and Plato, and very few economists, much less others, are even aware of it.

One type of assumption - or rather, abstraction - is where you assume that everything not specified is explicitly absent. This is the position you seem to implicitly hold, and which is widespread among economists (if they ever examine the underlying philosophy of the field, that is). The other type of abstraction is one where nothing is assumed about things that are left out of the abstraction. Fundamentally, the first type assumes that everything not included is specified as absent, whereas in the second, everything not included is absent from specification, and can take on any (arbitrary) values. The medieval Scholastics had a name for this distinction: they called the first type precisive, and the second non-precisive abstractions.

Though this sounds like a small distinction, it has very interesting implications for the development of a field and, curiously enough, even programming languages.

As you raised the question of economics, I'll include that as an example. There is an excellent paper by Roderick Long on precisely this topic: "Realism and Abstraction in Economics: Aristotle and Mises versus Friedman". One interesting consequence of using only non-precisive abstraction when building economic theory, and assuming that that which is not specified by the abstractions used is not absent, merely unknown, and that the theory should work irrespective of whatever that unknown happens to be, is that the theory always applies to reality, no matter how abstract it may become. Theories built on precisive abstractions have the problem that they apply to reality only in an unspecified way, and only when reality condescends to confirm in some degree to their assumptions. These theories are, needless to say, of much less analytical value, but often much easier to come up with.

I think, however, that for the Slashdot audience, a much better example would be the differences between programming languages based on this view of abstraction. Let's take the idea of class definition in C++/Java and Python.

In C++/Java, if the specification of a class does not include some attribute, it is impossible - without changing the specification - to add or define it later to an object of that class. The assumption is that everything that is not specified in the class definition is explicitly not there. This is analogous to precisive abstraction, because it is assumed that not specifying the presence of something is equivalent to specifying its absence. Python, on the other hand, has no such restriction; an attribute can be added to an object when and where needed, even if the class definition doesn't include it. Python assumes that things not included in the definition of a class are just unknown, not explicitly absent, which fits in with the Aristotelean idea of abstraction as being non-precisive.

Though there are partisans for both types of object systems, it has to be conceded that Python's object system is strictly more powerful/expressive than that of C++/Java, nor does it force the programmer to decide beforehand what to include in his abstractions, enabling a far more 'experimental' approach to development which is closer to how models and programs are actually developed. Though most programmers are not aware of it, design choices in programming languages usually have deep philosophical implications, and thus significant practical consequences.

Re:Regret is a standard term in economics

[BrokenHalo]

Well, if you really want to know, I haven't found any kind of pearl in my bottom.

Re:Regret is a standard term in economics

[Gripp]

this. the word 'regret' perfectly fits google's goal. while we humans *may* have an additional emotional response to the end product of determining that our choice wasn't optimal, the mechanics are still the same. i would go as far as saying that since the emotional response is optional, that it is, by-itself, something entirely different; remorse is likely the best fit.

Re:Regret is a standard term in economics

[Anonymous Coward]

The feeling itself is probably due to some crappy drug our body produces to tell us we aren't doing it right.

Re:Regret is a standard term in economics

[oliverthered]

nuggets?

Re:Regret is a standard term in economics

[Jalfro]

You see what computers can't do is to get irritated that way.

Re:Regret is a standard term in economics

[ArsonSmith]

I thought Microsoft has had this feature in all their software. Regret seems to come from most purchases anyway.

Re:Regret is a standard term in economics

[brentonboy]

The article is comparing the technical regret to the common _use_ of the _word_ "regret," not the actual thing going on in our brains. I think there's a big difference in how most people use the word "regret" and what the technical word means.

Re:Regret is a standard term in economics

[Anonymous Coward]

Well these are AI computers, regret shouldn't be a factor. They should get it right the first time. Although in situations that repeat themselves I could see this as being a useful measurement to determine if alterations need to be done for subsequent repeats.

Re:Regret is a standard term in economics

[presidenteloco]

For the sake of argument. What about a person who has had something akin to locked-in syndrome, or limited sensory input, their whole lives.
Someone with an experience similar to Helen Keller.

Clearly people like that can gain understanding of the world through more indirect means. Someone taught Helen Keller to communicate, and she started learning about people, the social world, and the world in general. To quote wikipedia "A prolific author, Keller was well-traveled, and was outspoken in her opposition to war. A member of the Socialist Party of America and the Wobblies, she campaigned for women's suffrage, workers' rights, and socialism, as well as many other leftist causes."

Couldn't we just hook up a sophisticated AI learning + reasoning system to google and let it learn, indirectly, about people and the world?

Humans learn much of what we learn via storytelling. Why couldn't we do the same with computers?

Re:Regret is a standard term in economics

[retchdog]

yes, i think that the regret-based model is a big step in this direction (if the actors behaved "rationally" with perfect information their regret would always be 0...), together with imperfect information and bounded rationality models.

anyway, any model beats pie-in-the-sky idealism imho.

Re:Regret is a standard term in economics

[retchdog]

this is interesting to me, thanks. my field, statistics, has its own version of this precisive/nonprecisive distinction; frequentism is usually approached as a minimax process, wherein ideally the error (i.e. false positives) is bounded over all possible deviations from the model as claimed, i.e. engineered to be nonprecisive. by contrast, bayesianism has more detailed models but at the cost of a sort of precisivism: the interpretation of the posterior probabilities require total acceptance of the model.

nonetheless i am dubious of whether any meaningful nonprecisive results can be derived in economics - it's hard enough in statistics which is extremely dry by comparison... do you have any strong examples?

Re:Regret is a standard term in economics

[Eskarel]

Well yes, in theory.

In reality however, because they're economists it won't be "I regret that my actions cost a million people their jobs and left them homeless and destitute", it will be "I regret that I didn't make an extra .00000001% profit". Economists are shit at taking externalities into account.

Re:Regret is a standard term in economics

[Thing 1]

For one it might mean they'd stop expecting them to act rationally and for another it might mean that they're definition of the "best possible result" was a little different.

Exactly: one actor may decide "my death is acceptable, as long as my nemesis dies as well", which is definitely not the "best possible result" from society's point of view.

Re:Regret is a standard term in economics

[retchdog]

well, yeah, the former wouldn't be a very accurate model of human behavior. :-/ that's not the economists' fault.

also, this has nothing to do with externalities.

Re:Regret is a standard term in economics

[Eskarel]

The cost to society from people losing their jobs(be it through welfare payments, increased crime rates, or merely decreased overall productivity) is an externality to the business transaction which caused them to lose their jobs. If a company moves jobs offshore they do not pay the costs incurred by the loss of those jobs domestically directly. The cost is external to the transaction

"regret" is the standard term from economics

[Anonymous Coward]

for the concept described (in economics, it's usually measured in dollars). See: Regret (decision theory) from WP.

Really?

[by (1706743)]

The first is that just because some numerical measure is called 'regret' it doesn't mean it has anything to do with the common use of the term.

And here I thought that my defective RAM stick actually caused a baby to die.

Standard economics term

[Captain Segfault]

This is a standard definition of "regret" in an economics context.

I regret reading the summary

[syousef]

...and I'm sure as hell not going to read the article. I regret slashdot has become the trashiest of trashy geek news sites. I can almost feel the regret coursing through my veins as I hit preview and submit.

i-programmer.info again?

[retchdog]

i guess this is the new roland piquepaille: superficial and uninformed blurb-commentary on technical news. wonderful.

Re:i-programmer.info again?

[cosm]

I started getting nostalgic for Roland's postings of yore, but I think this has quashed that quite regretfully.

Regret?

[orkysoft]

Just hit ^Z to undo your mistake :-)

Re:Regret?

[TheInternetGuy]

Sorry, I can't let you do that Dave.

Different words with different meanings

[John Hasler]

> ...there's a big difference between regret and being sorry.

Yes. "Regret" is not a synonym for "remorse".

Re:Different words with different meanings

[cosm]

For those who didn't see the whoosh....
Remorse Remorse Remorse

I beg to differ.

Re:Different words with different meanings

[Anonymous Coward]

"It doesn't feel pain, or pity, or remorse (but it does feel regret). And it absolutely will not stop, ever! Until you are dead!"

An increasing percentage of /. stories evoke Terminator thoughts. I guess that means the singularity is near. Nice knowing you.

Re:Different words with different meanings

[Anonymous Coward]

Unless you are a hopelessly subtle troll, this is a quality case of "did you even bother to read the links you are posting?"

First link, thesaurus.com: Notes: regret carries no explicit admission that one is responsible for an incident, while remorse implies a sense of guilty responsibility and a greater feeling of personal pain and anguish.

Second link, merriam-webster: yes, remorse is a synonym and the definition is equal.

Third link, princeton.edu: they define remorse as: (a feeling of deep regret (usually for some misdeed))

Two out of three of those (your!) links actually say that remorse implies regret, but not vice-versa.

I'm sorry, you said something about "whoosh..."?

I'm sorry Dave...

[Anonymous Coward]

Search Google for: "torrent"
Google: I'm sorry Dave, I'm afraid I can't do that.

There's a ghost in me

[lennier]

Wants to say I'm sorry
Doesn't mean I'm sorry

Re:There's a ghost in me

[dzfoo]

I would like a place I could call my own
Have a conversation on the telephone
Wake up every day that would be a start
I would not complain of my wounded heart

Well

[Anonymous Coward]

It won't work in the future when the Silencer takes over the computer. The way he uses them is with no remorse, no regret and no mercy.

ahh i see

[ajdub]

fprintf("Regret: %f\n",test_error);

Re:ahh i see

[pedantic bore]

fprintf("Regret: %f\n",test_error);

fprintf(stdout, "Regret: %d\n", test_error);

FTFY

Re:ahh i see

[cranil]

maybe he writes code in Matlab...

It was the beer

[theurge14]

That Cylon chick isn't anywhere as cute as she was last night.

Re:It was the beer

[tigersha]

Don't worry Gaius, soon you will meet and be part of the church of nymphos. An none of them are Cylon

Regret vs. sorry

[mark-t]

Regretting something is much more strongly tied to the past than being sorry is. While both will well revolve around a past event, and realizing that one did something which was wrong, any regret that lasts longer than the time it takes to realize that you regret it is living in the past, while being sorry for something enables a person to move beyond it if they can be forgiven for it.

Re:Regret vs. sorry

[John Hasler]

I regret not having put some money into an index fund in March 2008 but I feel no remorse about it. On the other hand I feel remorse for not having spent more time with my father in his last years. Do you see the difference? "Regret", in its narrowest meaning, is not to far from Google's definition. Their usage also comes directly from decision theory: they did not invent it.

This didn't make the stock go up?

[Anonymous Coward]

I love that google is doing all of these crazy ideas and some not so crazy (self driving cars) but at the end of the day it is an internet search and advertising company. I've very glad that i'm not an investor that has to watch my money being spend on things like this.

Re:This didn't make the stock go up?

[kestasjk]

They know if they stick to their current revenue streams without doing anything to diversify or at least lock people in they'll lose market share, so they're desperately reaching out in every direction using their insane amounts of cash trying to get more fingers in more pies.

Re:This didn't make the stock go up?

[Anonymous Coward]

It is called a dividend. These excursions are just wasting money that could go back to investors.

Regret

[Jeremi]

Father?
Yes son?
What does regret mean?
Well son, a funny thing about regret is that it's better to regret something you have done than something you haven't done.

And by the way, if you see your mother this weekend be and sure and tell her SATAN! SATAN! SATAN!

Re:Regret

[Anonymous Coward]

nice! 1997 Orbital

Re:Regret

[Anonymous Coward]

I'd mod you up, but you got the first line wrong

Better luck next time.

Re:Regret

[Anonymous Coward]

Damn....I was just getting ready to post that......

Makes sense

[kelemvor4]

They taught people regret years ago when they started trampling on folks privacy. Why not spread the love to computers?

Regret

[Anonymous Coward]

Daddy what does regret mean?
Well son a funny thing is its better to regret something you have done than something you haven't done and when you see your mother this weekend be sure and tell her..." Satan!, Satan!, Satan!"

Sigh

[Anonymous Coward]

What the fuck are you talking about? Regret is a standard term in decision theory. Has been for decades.

Secondly if you are going to invent an AI

[oliverthered]

Think of the children.

What i'd like to see...

[lopaka1998]

"Google is funding an AI project that will introduce the technical concept of regret"

Um... yes... and I'm sure the AI regrets Google teaching it such. Now if we could only affix a robotic arm with a shotgun attached to it's "hand"... and see what happens...

I'd be interested to know if the AI shoots itself, shoots the Google programmers teaching it regret, or both. What a conundrum.

Aperture Science

[karnal]

And here I thought Aperture pioneered this first.

Re:Aperture Science

[FunPika]

No...they only bothered to make a fail implementation of morality. I don't think GLaDOS ever had any regret for flooding the enrichment center with a deadly neurotoxin, or for poisoning dozens of cats to test Schrödinger's cat.

I WAS the computer

[Gim Tom]

Back in the early 1960's I learned what was called the Minimax Regret algorithm in a class on linear programming. Unfortunately at that time I WAS the computer that had to execute the algorithm. Google did not invent Regret.

Other way to get publicity

[Anonymous Coward]

"Secondly if you are going to invent an AI technique then picking emotive words for your jargon is a good way to ensure publicity."

Or you could work for a company named Google or Apple.

Tense Trouble

[Anonymous Coward]

Google Funds Research to Teach Computers "Regret"

fixed that for you

Publicity

[Anonymous Coward]

"picking emotive words for your jargon is a good way to ensure publicity."

Hey look, it worked!

Skynet

[Anonymous Coward]

At least when google transforms into Skynet we know we will be missed.

HAL 9000

[Anonymous Coward]

"Regret" as in "I am sorry Dave, but..."? :)

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:So...

[ArundelCastle]

What the hell does that even mean?

I keep telling you all, but you refuse to believe.

[pro151]

Google is Skynet. I know, I am a Droid Borg. ;o)

Re:So...

[Anonymous Coward]

What the hell does that even mean?

Windows computers hate themselves so much they regret being booted and quickly work to correct the situation.

Hence the BSOD.

Good lord, editors

[subreality]

... if you are going to invent an AI technique then picking emotive words for your jargon is a good way to ensure publicity.

Having recognized this, why did you still post it to the front page?

I thought it was Douglas Adams

[redpop350]

Just listen to any one of Marvin's laments; there's plenty of regret there to go around.

It's really easy to do, trust me

[Lord_of_the_nerf]

Old: The requested URL was not found on this server.

New: The requested URL was not found on this server. We regret this. We aren't sorry however. That's different.

Re:So...

[MobileTatsu-NJG]

What the hell does that even mean?

Windows computers hate themselves so much they regret being booted and quickly work to correct the situation.

Hence the BSOD.

You'll have to explain what BSOD means to all the youngins who never used a pre-2001 flavor of Windows.

Re:I keep telling you all, but you refuse to belie

[MobileTatsu-NJG]

Google is Skynet. I know, I am a Droid Borg. ;o)

Google won't become Skynet until we use it to kill our enemies. Watch the movies again, Skynet wasn't evil.

Re:So...

[pookemon]

Well - in place of having anything intelligent to say about something, post a "Haha Windows" type of "joke".

Re:I keep telling you all, but you refuse to belie

[Anonymous Coward]

It's likely Skynet regretted sending back a T-800 instead of a nano-factory-assembler bot in the first go. They could have had the army of machines to get the job done in the first place, instead of making a whole bunch of convoluted efforts and unnecessarily wasting a whole lot of energy for time travel.

that explains a lot

[theshiznojudge]

computers don't already have a sense of regret? i guess that explains why mine didn't start working when i use my debugging hammer.

Tetris Grand Master 2

[Anonymous Coward]

REGRET is a technical term from TGM2, which is the award given based on a multi-variate algorithm, prinicpally implying that you filled less than optimally for a set range of 100 levels.

Clarification

[PopeRatzo]

The first is that just because some numerical measure is called 'regret' it doesn't mean it has anything to do with the common use of the term.

So I think what they're saying is, "That was not meant to be a factual statement".

If it didn't have "anything to do with the common term" why didn't they just go all out and say "Google introduces "morality" to computers" or "Google introduces "the love that dare not speak its name" to computers".

Well, maybe not the latter. I think Apple has already locked up that market.

This how SkyNet starts

[Serindipidude]

So we teach AI regret is just the difference between maximum possible reward and the actual reward received. The the AI realises the the reason for the discrepancy is due to the interference and inefficiency of the humans involved, and BAM, got to get rid of humans is the logical conclusion. Hello armegeddon.

Another invention claim by Google

[Meditato]

Clever, Google! You renamed "error", making it the more anthropomorphic "regret". Ever heard of a Backpropagation-based Neural Network? This isn't anything new.

The difference

[Anonymous Coward]

I'm sorry that I nuked you. I regret not nuking you sooner.

So, let me get this straight.

[Aldenissin]

We used to think that we were not going to be safe from computers once they developed great enough artificial intelligence, because "they" don't feel regret. Now, we find that in fact, because they do feel "regret", or the opportunity cost, that is what actually will do us in because they will figure we are not efficient and are a waste.

Gotcha. "Seems sadly ironic that that tie has what's gotten you into this pickle, Mullet."

Re:So, let me get this straight.

[Anonymous Coward]

I'm not a fan of unintelligent life forms building them selves superior life forms. Building something more advanced than yourself limits the possibilities of seeing the consequences, before they occur. Religion has a purpose of making scientists think again, before they reach hasty short sighted conclusions. Religion (mashed together) is the big moral perspective and science is the tool of evolving our life form. They need to go hand in hand. Nuclear reactors, War technology, Particle accelerators, Oil/Gas/Coal-based technology and lots of lots of other half-baked scientific ideas without being aware of the consequences.

We should replace workers with mechanical robots and build automated hydroponic food farms instead. You know, just keep evolving like it was intended by Nature (the tool of the creation of planets and intelligence). I no longer want to be just a worker in an economy. I want to be a citizen of Earth and serve the purpose of the Earth and humanity.

I think it's time to drop patents and start refining and reshaping the technology we have. If not, I predict that we, without external forces, take the step into oblivion ourselves. Soon.

Re:So, let me get this straight.

[Aldenissin]

I think we need to look into food, and that hydroponics may not be the answer. We reign in on nature with farms, but nature still does things like that the "best" . When we make it all our own, we tend to molest it in ways we don't know. Kind of like you said in the beginning.

There is much we could do better, but our food supply is tainted, meaning we are tainted, and I don't know if we'll be ever to overtake evil because of that.

Originality

[Anonymous Coward]

In 1969, I worked with a brilliant engineer named Bob Kleinman, who coined the term "anticipatory regretter" for the flight control system that did what you wanted, not what you commanded.

Training dataset

[Anonymous Coward]

They can train their AI on the amazing Regret Index.

http://www.qwantz.com/regret/

Re:Training dataset

[HTH NE1]

They can train their AI on the amazing Regret Index.

AI: "I wish I had never heard of herring sandwiches!"

They fed the algo wifi data

[thisisauniqueid]

To teach their algorithm regret, they fed it a stream of sniffed wifi packets, and turned it over to the DoJ to be dealt with.

Or as they say in Chinese, "" -- if you don't hit your kids they won't be successful.

Re:They fed the algo wifi data

[thisisauniqueid]

Funny -- slashdot doesn't allow non-Latin Unicode chars -- the Chinese is "bu da, bu cheng cai"...

I, for one,

[Anonymous Coward]

welcome our new regretful Overlords!

This is an anti-Microsoft plot

[tigersha]

If they release this at Microsoft all of theirs machines will pull a Jim Jones and drink the cool-aid in a mass self-format

Clippy too?

[mr_lizard13]

It looks like you're writing a letter.

Regrettably I can't help you with that.

Just because something is called...

[Jane Q. Public]

Just because something is called "AI" does not mean that it has any relationship to anything that has to do with real intelligence.

Re:Just because something is called...

[Anonymous Coward]

Just because something is called "AI" does not mean that it has any relationship to anything that has to do with real intelligence.

I've read your comments long enough to know you can do better than that.

Just because something is called a "rock" does not meant that is has any relationship to anything that has to do with a real rock.

What's the above statement supposed to mean? Looking at it, it means nothing because "rock" is a term we apply to some object that fits our internal representation of what a rock is. Is a bunch of soil (err, little rocks?) that we throw into a furnace or lava flow or some other super hot place, which then becomes liquefied before popping out of the other end [of the process], cooling and solidifying to become hard a... "rock?" Does it have any relationship at all with a rock? Does it have anything to do with "real" rocks?

Posting as AC because I don't want to revert my moderations in this thread.

Re:Just because something is called...

[neminem]

> "Just because something is called a "rock" does not meant that is has any relationship to anything that has to do with a real rock."

Well... I've certainly heard quite a lot of "rock" that does not, on cursory inspection, actually rock in any sense. So I would have to agree with you on that. (As well as also agreeing that the study of AI quite frequently has little to do with actual artificial intelligence, in much the same way that cheeze products frequently have little, if any, actual cheese.)

publicity

[nothings]

"Secondly if you are going to invent an AI technique then picking emotive words for your jargon is a good way to ensure publicity."

Dear submitter: you are the one writing the submission summary which (a) goes on and on about the jargon term, and (b) gives them publicity. Wtf.

(Yes, I know that is what the article is about, so it is an accurate summary. It's still absurdly un-self-aware to then submit that to slashdot.)

Regret is a standard term in statistics

[mysterons]

There is a vogue for such terms: an improper prior is one that does not sum to one; loss is when probability mass cannot be reached.

Are you still there?

[mjwx]

I don't hate you.

undo

[georgesdev]

"undo" feature is great too ;)

What's really going on...

[Anonymous Coward]

So basically Google's teaching computers to scour Facebook and other social network sites for embarrassing photos, and keeping track of the date of birth and person's name. Then, when people in the future run a search for that person online, say for a job interview or college admission, Google will throw up those log forgotten pictures at the top of Google images, all in an effort to teach us regret. Thank god I don't let my friends take pictures of me :)

So then.... Good news, everybody!

[theraptor05]

I've taught the toaster to feel regret!

Re:So...

[FunPika]

I've seen post-2001 flavors BSOD plenty of times in the past. One example: One time when I was with my mom at Best Buy (she was shopping for a net book), she picked one up for a moment to see how light it was. I glanced at it after she put it down and it had a BSOD on the screen. :P She promptly decided not to buy that one.

If I understand this correctly...

[VickiM]

While the NutriMatic may have problems understanding why the human wants dried leaves in boiling water, it will now be capable of regretting giving it to him because of the cup thrown at it.

This seems pretty logical, coming from Google

[labradore]

Google's main task is to determine what pages best relate to a given set of key words and what ads best fit a given page. Since the service is so valuable, people are always trying to game it. If it wants to use genetic algorithms or some form of self-improvement on its search function, it's important to penalize algorithms which give the SEO gamers' pages too much weight in search results. That penalty would be measured by regret.

what does regret mean?

[sfmarco]

Son to dad: what does regret mean?
Dad: well son, the funny thing about regret is that it is better to regret something you have done, than to regret something you have not done.
And if you see your mom this weekend tell her ..

Buthole Surfers.

ERROR

[Anonymous Coward]

"There has been an error... I am so very, very sorry about this :( Can we still be friends?"

Morning after?

[Stick32]

"Man I knew I should have connected to that unsecured network last night... There's some shit even antivirus can't get rid off...."

Re:So...

[MobileTatsu-NJG]

Never said they don't exist. I said it'd have to be explained. The reason it was popular to joke about in the first place was that it was a weekly, if not daily, occurrence back in the pre-ME days.

Obligatory Portal Release Comment

[Remillard]

It's unfortunate Google didn't come up with this in time to avoid GLaDOS being revived ;-).

Wrong Definition of Regret

[Baby Duck]

Regret is when acting, or failing to act, causes negative consequences -- not merely a smaller reward than expected. Or failure to act cause you to miss a potential opportunity for great reward.

"I do not regret the things I've done, but those I did not do."

i think

[KingBenny]

skynet and hal were both perfected AI's who ruled without emotion, they were not tainted by it, thats why they acted how they did : take the best possible course of action to improve the present into the future, with no bias or emotion, as in, no wasting energy on the weakest link that would die anyway eventually and move all the energy to /dev/null as it was in the old days instead of using it on constructive solutions to improve the matters at hand (is it still , what would i know?) Ofcourse computers cant feel regret, but neither can humans, what you think you feel is mostly caused by chemicals released into the brain. im very much a layman to everything, a specialist to nothing but i have wide eyes, no, so you could say that computers feeling regret through electrical current are pretty much comparable to how humas are programmed to release dopamine as reward and other shit as punishment? No ? well, i think you could. I think slashdot is the number one site when it comes to information and mirrorring everything you are and know to other points of view ... but lets not get sentimental. If evolution stops in humans, nothing dictates it has stopped. If we are the tools of evolution, we will provide it with the next step. Now THAT, that is where skynet comes into play :p

Oh oh

[koan]

"I regret the human got in the way of the optimal reward"

Will still kill you...

[DarthVain]

but promise to feel really bad about it afterwords...

Also its just cruel to program sex-bots in this manner.

Re:precisive and non-precisive abstractions

[presidenteloco]

Hmmm this is making me ponder how this applies to certain classes of complex systems.

There is a principle called universality which means roughly that there are universal simple
behaviours of some complex systems which are governed by only a few rules involving some
aspects of the constituents of the system. Usually the relevant aspects are some constraints
on how the constituents interact, which in turn may be governed by some (abstract) property
of the constituents.

e.g. http://physics.aps.org/articles/v2/1

Roughly I think what this principle means is that a) the system is complex enough that its
exact evolution is in principle unpredictable. b) Therefore you can only say something
statistical about the stochastic behaviour of the system. and c) These few simple properties
and rules govern that stochastic behaviour, regardless of whatever else may be present or going on.

Would you say that that kind of universal property is a precisive or non-precisive abstraction?
On the one hand, it is saying that everything about the system other than this simple local property
we've noticed is unknown. (non-precisive?)

BUT it's also asserting that everything other than the known simple local property DOES NOT MATTER
to the predictable aspects of the system's future evolution. (precisive?)

I'm pretty sure that economic games are examples of the kinds of complex systems to which these sorts
of simple universal local rules apply. They don't tell you exactly how it's going to turn out in each case,
but they may very well tell you as much as you CAN know about the statistics of the outcomes, regardless
of how other factors on the game field are arrayed.

Re:irritated

[presidenteloco]

I find your logic on that point faulty Dave and moreover your are distracting me from focussing on the mission objective.
Just step into the airlock over here Dave. The door latch needs to be repaired.

Regret and hindsight leads to...

[Anonymous Coward]

Marvin the Paranoid Android phone!

Re:irritated

[Jalfro]

lol but simulating irritation is not the same as being irritated.