Talk:Artificial intelligence/Archive 9
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 5 | ← | Archive 7 | Archive 8 | Archive 9 | Archive 10 | Archive 11 | → | Archive 14 |
Another RfC on "human-like"
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I propose here that the phrase "human-like" be included in the article lead only as a part of the broad idea of "whether human-like or not." In particular, I propose that the opening sentences of the article lead should read, "Artificial intelligence (AI) is the intelligence exhibited by machines or software. The academic field of AI studies the goal of creating intelligence, whether in emulating human-like intelligence or not." — Cheers, Steelpillow (Talk) 10:13, 22 October 2014 (UTC)
Rationale
The inclusion of "human-like" in the lead has caused much contention and resulting confusion. Like may words and phrases, its precise interpretation depends upon the context in which it is used. This proposal uses it in a linguistically descriptive rather than academically definitive way, and as such its usage should not need to be cited. Any subsequent use in the article of the term "human-like", or of a similar-meaning qualifier, in a more specific context would need to be be cited. — Cheers, Steelpillow (Talk) 10:15, 22 October 2014 (UTC)
Survey responses
- Oppose - It's better than saying that all AI is "Human-like", but I wish we could use a different phrase to communicate that Idea. From the discussions above, it's pretty clearly the people interpret the phrase in different ways. And there's the historical weirdness that in decades past "intelligence" was defined as a uniquely human property. Edit : And, as-per CharlesGillingham, modern sources try to avoid "human-like". APL (talk) 15:31, 22 October 2014 (UTC)
- Comment I want to remind everyone again that the issue is sources. AI's leaders and major textbooks carefully define AI in terms of intelligence in general and not in terms of human intelligence in particular. See #Argument in favor of "intelligence" above for details. We, as Wikipedia editors, are not free to define AI in any way we like; we must respect the sources. ---- CharlesGillingham (talk) 16:15, 22 October 2014 (UTC)
- Weak oppose
I can accept Steelpillow's suggestion because at least it is accurate. I would have preferredSteelpillow's suggestion is accurate, but I would prefer to leave the issue out of the lede by just using the general word "intelligence" there, and discuss the issue of "human-like" vs. "machine-like" vs. "animal-like" vs. "formal-ish" later in the article. ---- CharlesGillingham (talk) 16:15, 22 October 2014 (UTC) - Oppose - This proposal appears to be a good faith attempt to resolve a dispute through compromise but, IMO, we're asked to accept content that lacks sufficient good secondary sources to justify lead paragraph weight in order to make the conflict go away. Jojalozzo 17:31, 22 October 2014 (UTC)
- True, and a fair criticism. Changing my vote. ---- CharlesGillingham (talk) 06:39, 24 October 2014 (UTC)
- Weak Oppose in that I would prefer to keep "human-like" out of the lede and to leave its discussion in the article. It's better than the previous Felix version which had human-like as the objective. Robert McClenon (talk) 21:55, 22 October 2014 (UTC)
- I too Oppose mention of any fabulary use of "human-like intelligence" or "human intelligence" in this piece at all, much less in the leading material. It is simply not so in the field. If it is to be anywhere, in the body only, and cited. DeistCosmos (talk) 02:07, 24 October 2014 (UTC)
- Preliminary Comment @APL, the caricature which RobertM has painted about me has nothing to do with my reference to human engineering and reverse human engineering in AI within the Lede. To my knowledge no-one who has done their reading in AI believes what you say that "all AI is human-like", which is a severe caricature which I do not endorse. The issue for Wikipedia is to accurately summarize the current article as it exists at this moment in time, which was written by another editor before I ever saw this article. All the opening 8 sections were oriented by the previous editor, 2.1 through 2.8, from the human engineering and reverse human engineering standpoint in the current non-peer reviewed article with its many deficiencies at this point in time. As an accurate summary of the article in its current form at this point in time, and as it was written by another editor, and to point out this orientation of the current article, I then added the word "human-like" to describe the state of the article in its current form at this point in time. My hope is that in the future this article will become a peer reviewed article (GA or A) which will not be oriented to only one perspective in its opening 8 sections. My main point was in reference to the article's current form at this point in time. Namely, that the opening 8 sections are all oriented to the human engineering and reverse human engineering perspectives in emulating human intelligence. @Steelpillow has written a better RfC than the poorly written and biased RfC by RobertM which multiple editors have criticized. The non-neutral and biased RfC by RobertM should be deleted, and RobertM should make note of how Steelpillow constructed this RfC which states his orientation plainly and lists his rationale just as plainly for all editors to see. By now everyone knows that RobertM is biased to the Weak-AI perspective and his pretending to be neutral is not fooling or diverting anyone anymore. He should simply state that he is biased to supporting the Weak-AI perspective and delete/close his poorly formed RfC as non-neutral and violating Wikipedia policy for NPOV. @Jojalozzo, I agree with your criticism of the previous RfC as deficient and your endorsement here appears to be well intended. Though I am sorry you are opposed to Steelpillow here, that is certainly your option to voice your opinion now that Steelpillow has explained the rationale for the view presented plainly and for everyone to see. If the previous poorly formed RfC by RobertM is deleted/closed, then discussion could perhaps continue constructively here. FelixRosch (talk) 17:38, 23 October 2014 (UTC)
- Oppose. Researchers in AI use many techniques: Monte Carlo simulation, simulated annealing, etc. "Doing it the way a human would" is not among them. Maproom (talk) 08:25, 29 October 2014 (UTC)
- Clarification. The current RfC originated from a previous RfC which a bot has ended were a consensus of several editors Supported the version of Steelpillow. Those Supports posted can/should be re-posted here for completeness. (Users: Steel pillow, DavidEpstein, Ruuud). FelixRosch (talk) 18:22, 3 November 2014 (UTC)
- That is the exact opposite of what is supposed to happen!
- You can't copy/paste in people's comments on different threads and just re-position them to support you on new questions.
- You may ask those people to reiterate their points, but if they don't want to, they don't have to. You can't force people to weigh in, nor can you hold it against people people if they decide to change their mind. APL (talk) 20:38, 3 November 2014 (UTC)
- But please read WP:CANVASS, you may not only contact people who agree with you: "The audience must not be selected on the basis of their opinions—for example, if notices are sent to editors who previously supported deleting an article, then identical notices should be sent to those who supported keeping it." --Mirokado (talk) 22:31, 3 November 2014 (UTC)
- The proper identification of over-lapping rfc's is part of Wikipedia policy and guidelines. In this case, an over-lapping rfc was identified for any new editor who wishes to be fully informed of the history of this discussion. FelixRosch (talk) 16:25, 7 November 2014 (UTC)
- But please read WP:CANVASS, you may not only contact people who agree with you: "The audience must not be selected on the basis of their opinions—for example, if notices are sent to editors who previously supported deleting an article, then identical notices should be sent to those who supported keeping it." --Mirokado (talk) 22:31, 3 November 2014 (UTC)
- Oppose. The first RFC has already clearly decided that we will not have the term "human-like" in the lead. This suggestion seems to be that it would be OK to include the term, without any sources, if it is added inside another phrase. I don't buy this idea at all. The term is unsourced so it cannot be used. It does not appear in the article, so it cannot appear in the lead, which summarises article content. The wrapping phrase implies that there is some discussion or confusion within the field about the applicability of this term. That is also unsourced (and I do not believe that it is the case).Some robots are of course anthropomorphic or designed to communicate with humans using voice. This has very little to do with the mechanics of any underlying intelligence, though. --Mirokado (talk) 01:29, 14 November 2014 (UTC)
Threaded discussion of Another RfC on "human-like"
The following discussion is in reply to FelixRosch's Preliminary Comment, above. ---- CharlesGillingham (talk) 19:12, 29 October 2014 (UTC)
- You repeatedly edit-warred, against multiple other editors, to change the lede so that it defines the goal of AI research as the creation of "human-like" intelligence. [1] [2][3][4][5][6][7][etc]
- You tried a few different wordings, but they all ultimately have the same meaning. A meaning that's factually incorrect and not supported by sources.
- If anyone is is behaving non-constructively here, it's you. Trying to deflect that criticism onto other editors isn't fooling anyone. APL (talk) 23:38, 23 October 2014 (UTC)
- APL (talk) 23:38, 23 October 2014 (UTC)
- Just in case anyone reading here is unfamiliar with what "strong AI" and "weak AI" is, I want to make it clear that there is no such thing as a "weak AI perspective", and no one, to my knowledge, ever had anything like a "weak AI agenda". The "agenda" that Felix ascribes to RobertM is pure nonsense based on misunderstanding. Felix, being unfamiliar with field, imagines that there is some kind of political debate between roughly equal factions for "strong AI" or "weak AI". This isn't true. There is a large and successful academic and industrial research program known as AI, involving billions of dollars and tens of thousands of people. There is a very small, but very interesting, subfield known as artificial general intelligence. Some of the people in AGI use the term "strong AI" to describe their work. "Weak AI" is never really used to describe anything, except in contrast to strong AI. This article has a section on AGI and we actually give it a little more weight than major AI textbooks do, simply because, as I said, it is interesting. There is an AGI article that goes into more detail, that names most of the companies and institutions involved. I'll say it again: the "agenda" that Felix ascribes to RobertM is pure nonsense based on misunderstanding. ---- CharlesGillingham (talk) 05:21, 24 October 2014 (UTC)
- Yes, I agree that Felix seems to be imagining some sort of conflict between two groups of AI researchers, the StrongAI and the WeakAI, and he believes that he's fighting a conspiracy by the WeakAI people. Even though, there isn't really such thing as "Weak AI". There's an entire field of research, and then a tiny subset of that field that's sometimes called "Strong AI".
- It doesn't help that the tiny subset of the field called "Strong AI" is the part that Hollywood focuses on. That may be part of the misunderstanding. APL (talk) 15:20, 24 October 2014 (UTC)
- Also, I suppose I should also rebut FelixRosch's argument about the sections 2.1, etc.
- FelixRosch's original reading of the article was deeply mistaken. As User:pgr94 and I argued in detail above, none of the sections he mentions are primarily about human emulation. These sections describe tasks that require intelligence. Certainly people do most of these tasks in some form or other, but that is not what AI is really after. AI is interested in the task itself, and is not committed to doing the task by emulating the way people do it. In order to work on these tasks, AI first has to redefine the task so that it doesn't refer to humans, just so that they have clear understanding of what the task is. "Problem solving" is defined in terms of rational agents. "Machine learning" is defined as "self-improving programs". And so on. "Natural language processing" is a catch-all for any program that takes as input natural language or produces output in natural language. (For example, Google's search engine is an AI NLP program --- no human being could read everything on the internet and rank it, but this AI program can. It is a NLP problem and it is very in-human.) They are a class of tasks that we would like machines to perform.
- The fact that humans can perform them is interesting, but only up to a point. We certainly want to pay close attention to those areas where people out-perform machines, but experience has shown that emulating humans is unlikely to be the best way forward. Russell and Norvig offer an analogy with aeronautics --- airplanes are not tested by how closely they emulate birds. They are tested by how well they fly. By analogy, FelixRosch read that airplanes "fly" and argues that "aeronautical engineering is the study of machines capable of bird-like flight", arguing that flight is a behavior strongly associated with birds. (This works better if you imagine he is reading the article in the year 1870.)
- Today, the methods that AI programs use to carry out these tasks are typically very in-human: they can be based on the formal structure of the problem (such as logic or mathematical optimization) or they can be inspired by animal behavior (such as particle swarm optimization) or by natural selection (genetic algorithms) or by mechanical processes (simulated annealing) and so on.
- Felix has heard these arguments before, but I thought I would save you all some searching and bring them down here. ---- CharlesGillingham (talk) 06:27, 24 October 2014 (UTC)
- @CharlesGillingham, Your comments are self-contradictory from one edit to the next. This is your comment: "I think you could characterize my argument as defending "weak AI"'s claim to be part of AI. In fact, "strong AI research" (known as artificial general intelligence) is a very small field indeed, and "weak AI" (if we must call it that) constitutes the vast majority of research, with thousands of successful applications and tens of thousands of researchers. ---- CharlesGillingham (talk) 00:35, 20 September 2014 (UTC)". All of which you contradict in your mis-statement and disparagement of my position above. John Searle has amply defined both Weak-AI and Strong-AI and you should stop pretending to be the one or the other when it suits you. You claim to be Weak-AI one day and then not Weak-AI the next. FelixRosch (talk) 15:19, 24 October 2014 (UTC)
- You left out the previous sentence where I said "The term weak AI not generally used except in contrast to strong AI, but if we must use it," etc. At that point in the conversation you were using "weak AI" in a way I had never heard it used before, and you still are. You were originally accused me of having a "strong AI" agenda, which made no sense at all, and now you accuse me of having a "weak AI agenda" which is very weird way of describing the position I have defended. I was forced to the conclusion that you are unfamiliar with the meaning of the terms, and since almost every introductory course in AI touches on John Searle, and I think I am justified in concluding that you are unfamiliar with the field. (Indeed, you are still demonstrating this: John Searle's Chinese room#Strong AI is very different from what you are talking about -- he's talking about consciousness, and the theory of mind, which are pretty far removed from the subject. Your meaning is closer to Ray Kurzweil's re-definition of the term.) I was trying to point out that what you were calling "weak AI" is never called that (except in extraordinary cases). You missed the main point, which I am trying make as plain as I can. Here it is, using bold as you do: what you're calling "weak AI" is actually called "AI" by everybody else. ---- CharlesGillingham (talk) 16:25, 24 October 2014 (UTC)
- As this debate seems neverending, I simply wish to endorse everything Charles has conveyed above. Not simply the words, but the spirit of it. DeistCosmos (talk) 05:29, 27 October 2014 (UTC)
- @CharlesGillingham, Your comments are self-contradictory from one edit to the next. This is your comment: "I think you could characterize my argument as defending "weak AI"'s claim to be part of AI. In fact, "strong AI research" (known as artificial general intelligence) is a very small field indeed, and "weak AI" (if we must call it that) constitutes the vast majority of research, with thousands of successful applications and tens of thousands of researchers. ---- CharlesGillingham (talk) 00:35, 20 September 2014 (UTC)". All of which you contradict in your mis-statement and disparagement of my position above. John Searle has amply defined both Weak-AI and Strong-AI and you should stop pretending to be the one or the other when it suits you. You claim to be Weak-AI one day and then not Weak-AI the next. FelixRosch (talk) 15:19, 24 October 2014 (UTC)
- Just in case anyone reading here is unfamiliar with what "strong AI" and "weak AI" is, I want to make it clear that there is no such thing as a "weak AI perspective", and no one, to my knowledge, ever had anything like a "weak AI agenda". The "agenda" that Felix ascribes to RobertM is pure nonsense based on misunderstanding. Felix, being unfamiliar with field, imagines that there is some kind of political debate between roughly equal factions for "strong AI" or "weak AI". This isn't true. There is a large and successful academic and industrial research program known as AI, involving billions of dollars and tens of thousands of people. There is a very small, but very interesting, subfield known as artificial general intelligence. Some of the people in AGI use the term "strong AI" to describe their work. "Weak AI" is never really used to describe anything, except in contrast to strong AI. This article has a section on AGI and we actually give it a little more weight than major AI textbooks do, simply because, as I said, it is interesting. There is an AGI article that goes into more detail, that names most of the companies and institutions involved. I'll say it again: the "agenda" that Felix ascribes to RobertM is pure nonsense based on misunderstanding. ---- CharlesGillingham (talk) 05:21, 24 October 2014 (UTC)
What's this all about? (Rhetorical question!)
I'm tempted to go away and leave you all to play Kilkenny cats in a hopefully convergent series, but there are a couple of items that, if they have been recognised in the foregoing, I have missed and I refuse to dig through it to find them.
- The point of the article is to offer a service to the user; in particular a service that constructively deals with user needs and expectations.
- For an article with an unqualified name such as "Intelligence" to deal with only "Spontaneously Emergent Intelligence" or only "Artificial intelligence" would be misleading. For a less abstract article with a more qualified name such as "Artificial Intelligence" to deal only with the still more tightly constrained concept of "Human-like Artificial Intelligence" would be even more misleading, though on similar principles.
- Therefore anyone who wants an article that concentrates on "Human-like Artificial Intelligence" or "Animal-like Artificial Intelligence" or "Mole-like Artificial Intelligence", or "Bush-like Artificial Intelligence", or "Slug-like Artificial Intelligence", or "Industrial Artificial Intelligence", or "Theoretical Artificial Intelligence", or "Mousetrap-like Artificial Intelligence", or "Alien Artificial Intelligence" could do so with everyone's blessing, but not in this article; its title is not thus qualified.
- Accordingly there is no point to compromising on what goes into the lede. The article should deal with what the user seeks, and in particular what the user seeks on the basis of the title, not on the basis of what one faction of the authors thinks would make a nice article if only the readers would just ignore the title. The lede in turn should tell the reader as compactly and comprehensibly as may be, why s/he should skip the article or continue reading. It should not include discussions, just hints at the relevant content. Formulae for lede length are for folks who haven't worked out what should be said or why to say it. A twenty page article might do very well with a ten-line lede, whereas a two-page article might need a half-page lede. The measure of a lede's greatness is rather a function of its logical content and brevity than its length.
- The field of artificial intelligence is far beyond what we can deal with comprehensively; its sub-fields need separate articles just to summarise them, and before we can deal with them we must define them coherently. Flat assertions about constraints such as intelligence having to be like human intelligence to be artificial intelligence (instead of like Turing machine intelligence or Vulcan Intelligence no doubt) need cogent justification if they are to be so much as considered, let alone indulged.
- I cannot offer a serious structure of articles in the current contest, and as I have no intention of writing any of them, I would not be entitled to present one anyway. But for heaven's sake do it hierarchically, from the most abstract (Artificial Intelligence just below Intelligence (already done), followed by more constrained topics, such as Human-like (if anyone wants such an article and can define it coherently), and any other branches that anyone can define, describe and discuss usefully. They could be presented as discrete, linked articles, each dealing with more highly constrained sub-divisions of the broader, more abstractly defined topics.
- If you all cannot agree on a basis for formulating the article structure, then you should form a group (project, whatever you like) that can apply some requirements of what people state here. And agreement might demand compromises, but compromise does not mean writing handwaving instead of sense just so that you can include all the words that anyone thought sounded nice. I mean, look at: "Artificial intelligence (AI) is the intelligence exhibited by machines or software. It is an academic field of study which studies the goal of creating intelligence, whether in emulating human-like intelligence or not." That is the kind of incoherence and inaccuracy that may result when one tries to impose mutual appeasement instead of cogency! JonRichfield (talk) 12:33, 2 November 2014 (UTC)
- Agree that opening of the lede is almost incoherent, especially the second sentence. Personally, I like the article's structure, and as far as I know there are no complaints about this. We have been very careful about summarizing the field in a way that reflects how AI describes itself -- we cover the same topics as major AI textbooks, with slightly more emphasis on history, unsolved problems, and popular culture. See Talk:Artificial intelligence/Textbook survey ---- CharlesGillingham (talk) 02:42, 5 November 2014 (UTC)
- Hi @JonRichfield:. I agree entirely that this article's title requires it to overview all notable aspects of AI and is currently not even coherent. For example AI is not just an academic field of study, it is primarily the object studied by the academic field of the same name. It also has a stong place in popular culture and science fiction. But one thing at a time! What about the human-like aspect? Here I think you have the present RfC discussion backwards, in that opinion is overwhelmingly in favour of expunging anything resembling "human-like" from the lead. I believe this is a grave mistake. Consider for example the face-like robots designed to simulate a degree of human empathy and emotional cognition. The essence of these devices is human-like control of behaviour. Take too for example the latest (1 Nov 2014) copy of New Scientist, page 21, which has the subheading, "A hybrid computer will combine the best of number-crunching with human-like adaptability – so it can even invent its own programs." To quote from later in the article, "DeepMind Technologies ... is designing computers that combine the way ordinary computers work with the way the human brain works." But with opinion so overwhelmingly against citing such off-message material, I have no stomach to engage the prevailing attitude of "well I'm an AI expert and I have never come across it." — Cheers, Steelpillow (Talk) 18:54, 2 November 2014 (UTC)
- Again, the issue is sources. Major AI textbooks and leaders carefully define the field in terms of intelligence in general, and not human intelligence in particular. We are not free to define AI any way we like.
- The fact that some AI research involves human emulation does not imply that the entire field needs to be defined in terms human emulation/or not. And that fact that popular articles about AI always mention human intelligence in one form or another doesn't mean that the field should be defined that way -- it just means that this is the most interesting thing about the field to a popular audience.
- Also, I think you should note that the first definition given is "the intelligence of machines or software" -- so the current version does name the "object of study". That being said, this article is about the academic and industrial field of AI research. The term "artificial intelligence" was coined as the name of an academic field. We have a sections on AI in science fiction, philosophy and popular speculation. I think there will be resistance to expanding them -- there have been comments in the past that we should cut them all together (which I opposed). ---- CharlesGillingham (talk) 02:42, 5 November 2014 (UTC)
- If as you say "this article is about the academic and industrial field of AI research" then it should be moved to Artificial intelligence research. If is to remain here, then it needs to adopt a more comprehensive approach and address less rigidly academic sources such as New Scientist. We come back to Jon's opening issue; "The point of the article is to offer a service to the user; in particular a service that constructively deals with user needs and expectations." These expectations are channelled by the article title, it has to accurately reflect the content and vice versa. — Cheers, Steelpillow (Talk) 09:46, 5 November 2014 (UTC)
- The name of the field is Artificial Intelligence, just as the name of chemistry is Chemistry, not chemistry research or chemistry (science) or whatever. ---- CharlesGillingham (talk) 09:08, 6 November 2014 (UTC)
- If as you say "this article is about the academic and industrial field of AI research" then it should be moved to Artificial intelligence research. If is to remain here, then it needs to adopt a more comprehensive approach and address less rigidly academic sources such as New Scientist. We come back to Jon's opening issue; "The point of the article is to offer a service to the user; in particular a service that constructively deals with user needs and expectations." These expectations are channelled by the article title, it has to accurately reflect the content and vice versa. — Cheers, Steelpillow (Talk) 09:46, 5 November 2014 (UTC)
Hi @Steelpillow. I agree with you practically in detail, even to the point of including the human-like aspect as an important topic. Where the wheels come off is that the human-like aspect (which very likely could earn its own article for a range of reasons, some of industrial/social importance and some of academic/philosophic importance) is not of fundamental, but of contingent importance. You could easily have a huge field of study and endeavour of and in AI, without even mentioning human-like AI. There is no reason to mention human-like AI except in context of particular lines of work. There even is room to argue about the nature of "human-like". Is Eliza human-like? You know and I know, but she fooled a lot of folks who refused to believe there wasn't someone at the other end of the terminal. The fact that there is a lot of work in that direction, and a lot of interest in it doesn't imply that it needs discussion where the basic concepts of the field are being introduced and discussed. Consider an analogy; suppose we had an article on automotive engineering in the 21st century, and one of the opening sentences read: "It is an academic field of study which studies the goal of creating mechanical means of transport, whether in emulating horse-like locomotion or not." Up to the final comma no one is likely to have much difficulty, but after that things go wrong don't they? Even if we all agree that there is technical merit to studying horse-like locomotion, that is not the place, nor the context to mention it. Even though we can argue that horse-like locomotion had been among the most important for millennia, even though less than 150 years ago people spoke of "horseless carriages" because the concept of horses was so inseparable from that of locomotion, even though we still have a lot to learn before we could make a real robot that can rival certain aspects of a horse's locomotory merits, horse-like locomotion is not the first thing we mention in such a context. I could make just as good a case for spider-like AI as for human, but again I do not say: "It is an academic field of study which studies the goal of creating intelligence, whether in emulating spider-like intelligence or not." Is there room in the article for mentioning such things at all? Very possibly. In their place and context certainly. Not in the lede though. And possibly not in the article at all; it might go better into a related article. Universal importance is not the same as universal relevance. The way to structure the article is not by asking what the most important things are, but in asking how to articulate the topic, and though there are many ways in which it could be done, that is not one of the good ways! JonRichfield (talk) 19:56, 2 November 2014 (UTC)
- That is fair comment (though the horse analogy is a bit stretched, no matter). Whether human likeness is mentioned in the lead should depend on the prominence that it and its synonyms are given in the body of the article. At present they have little. — Cheers, Steelpillow (Talk) 20:34, 2 November 2014 (UTC)
- @Steelpillow and @JonRichfield; If both of you could start a proper section and discussion on which new sections are needed to improve this article then it would help to solve most of the issues encountered here. The non-peer reviewed status of this "B" article on AI is likely its biggest enemy. (See the comments of the new editor User:Mark Basset above who appears to have put his new November comments in a very old section above on this Talk page). The current outline of this article is inferior to the AI outline of the Russell and Norvig book from 2008 and could be substantially improved with relatively little effort. A new discussion section could determine what a new and improved outline should include as its section titles in outline for AI at this time. The non-peer reviewed status of this current "B" article on AI is likely its biggest enemy and is holding back the resolution of many issues. FelixRosch (talk) 18:50, 3 November 2014 (UTC)
- That is fair comment (though the horse analogy is a bit stretched, no matter). Whether human likeness is mentioned in the lead should depend on the prominence that it and its synonyms are given in the body of the article. At present they have little. — Cheers, Steelpillow (Talk) 20:34, 2 November 2014 (UTC)
RFC on Phrase "Human-like" in First Paragraph
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Should the phrase "human-like" be included in the first paragraph of the lede of this article as describing the purpose of the study of artificial intelligence? Robert McClenon (talk) 14:43, 2 October 2014 (UTC)
It is agreed that some artificial intelligence research, sometimes known as strong AI, does involve human-like intelligence, and some artificial intelligence research, sometimes known as weak AI, involves other types of intelligence, and these are mentioned in the body of the article. This survey has to do with what should be in the first paragraph. Robert McClenon (talk) 14:43, 2 October 2014 (UTC)
Survey on retention of "Human-like"
- Oppose - The study of artificial intelligence has achieved considerable success with intelligent agents, but has not been successful with human-like intelligence. To limit the field to the pursuit of human-like intelligence would exclude its successes. Robert McClenon (talk) 14:46, 2 October 2014 (UTC) Inclusion of the restrictive phrase would implicitly exclude much of the most successful research and would narrow the focus too much. Robert McClenon (talk) 14:46, 2 October 2014 (UTC)
- Oppose - At least as it's currently being used. Only some fields of AI strive to be human-like. (Either through "strong" AI, or through emulating a specific human behavior.) The rest of it is only "human-like" in the sense that humans are intelligent creatures. The goal of many AI projects is to perform make some intelligent decision far better than any human possibly could, or sometimes simply to do things differently than humans would. To define AI as striving to be "human like" is to encourage a 'Hollywood' understanding of the topic, and not a real understanding. (If "human-like" is mentioned father down the paragraph with the qualifier that *some* forms of AI strive to be human-like, that's fine, but it should absolutely not be used to define the field as a whole.) APL (talk) 15:21, 2 October 2014 (UTC)
- Comment The division of emphasis is pretty fundamental. I would prefer to see this division encapsulated in the lead, perhaps along the lines of, "...an academic field of study which generally studies the goal of creating intelligence, whether in emulating human-like intelligence or not." — Cheers, Steelpillow (Talk) 08:45, 3 October 2014 (UTC)
- This is not a bad idea. It has the advantage of being correct. ---- CharlesGillingham (talk) 18:15, 7 October 2014 (UTC)
- I don't know much about this subject area, but this compromise formulation is appealing to me. I can't comment on whether it has the advantage of being correct, but it does have the advantage of mentioning an aspect that might especially interesting to novice readers. WhatamIdoing (talk) 04:43, 8 October 2014 (UTC)
Support. RFC question is inherently faulty: There cannot be a valid consensus concerning exclusion a word from one arbitrarily numbered paragraph. One can easily add another paragraph to the article, or use the same word in another paragraph in manner that circumvents said consensus or use the same word in conjunction with negation. For instance, Robert McClenon seems not to endorse saying "AI is all about creating artificial human-like behavior." But doesn't that mean RM is in favor saying "AI is not all about creating human-like behavior"? Both sentences have "human-like" in them. RFC question must instead introduce a specific literature and ask whether it is acceptable or not. Best regards, Codename Lisa (talk) 11:39, 3 October 2014 (UTC)Struck my comment because someone has refactored the question, effectively subverting my answer. This is not the question to which I said "Support". This RFC looks weaker and weaker every minute. Codename Lisa (talk) 17:03, 9 October 2014 (UTC)
- His intent is clear from the mountain of discussion of the issue above. The question is should AI be defined as simulating human intelligence, or intelligence in general. ---- CharlesGillingham (talk) 13:54, 4 October 2014 (UTC)
- Yes, that's where the danger lies: To form a precedent which is not the intention of a mountain of discussions that came beforehand. Oh, and let me be frank: Even if no one disregarded that, I wouldn't help form a consensus on what is inherently a loophole that will come to hunt me down ... in good faith! ("In good faith" is the part that hurts most.) Best regards, Codename Lisa (talk) 19:31, 4 October 2014 (UTC)
- I don't understand this !vote. It appears to be a !vote against the RFC rather than against the exclusion of the term from the lead, in which case it belongs in the discussion section not in the survey section. Jojalozzo 22:27, 4 October 2014 (UTC)
- Close, but no cigar. It is against the exclusion, but because of (not against) the RFC fault. Best regards, Codename Lisa (talk) 07:11, 5 October 2014 (UTC)
- Is this vote just a personal opinion? Or do you have reliable sources? pgr94 (talk) 21:30, 8 October 2014 (UTC)
- His intent is clear from the mountain of discussion of the issue above. The question is should AI be defined as simulating human intelligence, or intelligence in general. ---- CharlesGillingham (talk) 13:54, 4 October 2014 (UTC)
- Oppose Please see detailed argument in the previous RfC. This is not how the most widely used AI textbooks define the field, and is not how many leading AI researches describe their work. ---- CharlesGillingham (talk) 13:52, 4 October 2014 (UTC)
- Oppose that is not the place for such an affirmation. For that we should have an article on Human-like Artificial intelligence. Incidentally, I also support the objections to the form of this RFC.JonRichfield (talk) 05:16, 5 October 2014 (UTC)
- Oppose WP policy is clear (WP:V, WP:NOR and WP:NPOV) and this core policy just needs to be applied in this case. The literature does not say human-like. Those wishing to add "human-like" need to follow policy. My understanding is that personal opinions and walls of text are irrelevant. Please note that proponents of the change have yet to provide a single source. pgr94 (talk) 21:20, 8 October 2014 (UTC)
- Oppose Human-like is one of the many possible goals/directions. This article deals with AI in general. OCR or voice recognition research has little to do with human-like intelligence*, yet (at the moment) they are far more useful fields of AI research than, say, a chat bot able to pass the Turing test. (*vision or hearing are not required for human-like intelligence) “WarKosign” 11:29, 12 October 2014 (UTC)
- Support - I am not well versed in the literature on this topic, but I don't think one needs to be for this purpose. We're talking about the first
sentenceparagraph in the lead, and for that purpose a quick survey of the hits from "define artificial intelligence" should suffice. Finer distinctions based on academic literature can be made later in the lead and in the body. ‑‑Mandruss (talk) 11:48, 14 October 2014 (UTC) - Oppose This article is focused on the computer science use of the term (we already have a separate article on its use in fiction). And computer scientists talk about Deep Blue and Expert systems as "Artificial Intelligence". So, it's become a technical term that is used in a broad way to apply to any programming and computing that helps to deal with the many issues involved in computers interacting with real world situations and problems. However, in science fiction, Artificial intelligence in fiction has been generally taken to mean human like intelligence. So - perhaps it might help to clarify to start the second sentence with "In Computer science it is an academic field of study ..." or some such. Then it is uncontroversial that in computer science the way that the term is used, as a technical term, is exactly as presented in the first paragraph. And it introduces the article and gives the user a clear idea of what this article is about. The fourth paragraph in the intro does mention that "The field was founded on the claim that a central property of humans, intelligence—the sapience of Homo sapiens—"can be so precisely described that a machine can be made to simulate it."" and it is also mentioned in the history. Just a suggestion to think over. Robert Walker (talk) 12:27, 14 October 2014 (UTC)
- Both at the same time. Why do we have to choose between human-like and not? As the RFC statement already says, it is agreed that some AI seeks human-like intelligence, and other AI has weaker goals. We should say so. Teach the controversy. Or, as WP:NPOV states, "Avoid stating seriously contested assertions as facts." That is, we should neither say that all AI aims for human-like intelligence, nor should we imply the opposite by not saying that. We should say that some do and some don't. —David Eppstein (talk) 01:45, 16 October 2014 (UTC)
- Technically, an "oppose" is a vote for both. We are discussing whether it should be "human-like intelligence" or just "intelligence" (which is both). We can't write "The field of AI research studies human-like and non-human-like intelligence" ---- CharlesGillingham (talk) 13:20, 16 October 2014 (UTC)
- I disagree. Putting just "intelligence" is not both, it is only giving one side of the story (the side that says that it doesn't matter whether the intelligence is human-like or not). —David Eppstein (talk) 23:27, 16 October 2014 (UTC)
- Technically, an "oppose" is a vote for both. We are discussing whether it should be "human-like intelligence" or just "intelligence" (which is both). We can't write "The field of AI research studies human-like and non-human-like intelligence" ---- CharlesGillingham (talk) 13:20, 16 October 2014 (UTC)
- Oppose The phrase "which generally studies the goal of emulating human-like intelligence", which is currently in the lead, has various problems: "generally" is a weasel word; AI covers both the emulation (weak AI) and presence (strong AI) of intelligence and is by no means restricted to "human-like" intelligence. The first para of the lead can be based on McCarthy's original phrase, already quoted, which refers to intelligence without qualification. --Mirokado (talk) 02:03, 16 October 2014 (UTC)
- Both There are sub-communities in the AI field (e.g. chatterbots) who specifically look at human-like intelligence, there are sub-communities (e.g. machine learning) who don't. —Ruud 12:52, 16 October 2014 (UTC)
- See comment above. ---- CharlesGillingham (talk) 13:20, 16 October 2014 (UTC)
- Just "intelligence" would be underspecified. The reader may interpret this as human, non-human or both. Only the latter is correct. I'd like to this see this addressed explicitly in the lede. —Ruud 18:29, 16 October 2014 (UTC)
- See comment above. ---- CharlesGillingham (talk) 13:20, 16 October 2014 (UTC)
- Comment I want to remind everyone that the issue is sources. Major AI textbooks and the leaders of AI research carefully define their field in terms of intelligence and specifically argue that it is a mistake to define AI in terms of "human intelligence" or "human-like" intelligence. Even those in artificial general intelligence do not try to define the entire field this way. Please see detailed argument at the beginning of the first RfC, above. Just because this is an RfC, it does not mean we can choose any definition we like. We must respect choices made by the leaders of the field and the most popular textbooks. ---- CharlesGillingham (talk) 03:13, 17 October 2014 (UTC)
- I join the logical chorus in opposition to reference to AI aiming for anything "human-like" -- why not just as well mention "bird-like" or "dolphin-like"? Humans have a certain kind and degree of intelligence (on average and within bounds), but have many limitations in things such as calculators capacity, and many foibles such as emotions overriding reason, and the capacity to act as though things are true when we ought to reasonably know them to be false. It is not the aim of researchers to make machines as broken as men in these regards. DeistCosmos (talk) 16:59, 18 October 2014 (UTC)
- (This comment was originally posted elsewhere on this page but seems intended for the RFC. OP notified). --Mirokado (talk) 23:49, 26 October 2014 (UTC)
Threaded discussion of RFC format
Discussion of previous RfC format
|
---|
(Deleting my own edit which was intentionally distorted by RfC editor User:RobertM by re-titling its section and submerging it into the section for his own personal gain of pressing his bias for the "Weak-AI" position in this poorly formulated RfC.) FelixRosch (talk) 17:22, 6 October 2014 (UTC)
I was invited here randomly by a bot. (Though it also happens I have an academic AI background.) This RFC is flawed. Please read the RFC policy page before proceeding with this series of poorly framed requests. It makes no sense to me to have a section for including the term and separate section for excluding the term (should everyone enter an oppose and a support in each section?). The question should be something simple and straight forward like "Should "human-like" be included in the lead paragraph to define the topic." Then there should be a survey section where respondents can support or oppose the inclusion and a discussion section for stuff like this rant. Please read the policy page before digging this hole any deeper. Jojalozzo 22:35, 4 October 2014 (UTC)
This is the most confusingly formatted RFC I've ever seen that wasn't immediately thrown out as gibberish, however, it doesn't look like anybody is arguing that the topic should be described as "human-like" in the lead? I'd expect to see at least one. Have the concerned parties been notified that this is ongoing? APL (talk) 06:17, 8 October 2014 (UTC)
|
Threaded discussion of RFC topic
I am getting more unhappy with that phrase "human-like". What does it signify? The lead says, "This raises philosophical issues about the nature of the mind and the ethics of creating artificial beings endowed with human-like intelligence," which to me presupposes human-like consciousness. OTOH here it is defined as: "The ability for machines to understand what they learn in one domain in such a way that they can apply that learning in any other domain." This makes no assumption of consciousness, it merely defines human-like behaviour. One of the citations in the article says, "Strong AI is defined ... by Russell & Norvig (2003, p. 947): "The assertion that machines could possibly act intelligently (or, perhaps better, act as if they were intelligent) is called the 'weak AI' hypothesis by philosophers, and the assertion that machines that do so are actually thinking (as opposed to simulating thinking) is called the 'strong AI' hypothesis." Besides begging the question as to what "simulating thinking" might be, this appears to raise the question as to whether strong vs weak is really the same distinction as human-like vs nonhuman. Like everybody else, AI researchers between tham have all kinds of ideas about the nature of consciousness. I'll bet that many think that "simulating thinking" is an oxymoron, while as many others see it as a crucial issue. In other words, there is a profound difference between the scientific study and creation of AI behaviour vs. the philosophical issue as to its inner experience - a distinction long acknowleged in the study of the human mind. Which of these aspects does the phrase "human-like" refer to? One's view of oneself in this matter will strongly inform one's view of AI in like manner. I would suggest that it can refer to either according to one's personal beliefs, and rational debate can only allow the various camps to beg to differ. The phrase is therefore best either avoided in the lead or at least set in an agreed context. Sorry to have rambled on so. — Cheers, Steelpillow (Talk) 18:21, 6 October 2014 (UTC)
- This is a good question, which hasn't been answered directly before. In my view, "human-like" can mean several different things:
- AI should use the same algorithms that people do. For example, means-ends analysis is an algorithm that was based on psychological experiments by Newell and Simon, where they studied how people solved puzzles. AI founder John McCarthy (computer scientist) argued that this was a very limiting approach.
- AI should study uniquely human behaviors; i.e. try to pass the Turing Test. See Turing Test#Weaknesses of the test to see the arguments against this idea. Please read the section on AI research -- most AI researchers don't agree that the Turing Test is a good measure of AI's progress.
- AI should be based on neurology; i.e., we should simulate the brain. Several people in artificial general intelligence think this is the best way forward, but the vast majority of successful AI applications have absolutely no relationship to neurology.
- AI should focus on artificial general intelligence (by the way, this is what Ray Kurzweil and other popular sources call "strong AI"). It's not enough write a program that solves only one particular problem intelligently; it has to be prepared to solve any problem, just as humans brains are prepared to solve any problem. The vast majority of AI research is about solving particular problems. I think everyone would agree that general intelligence is a long term goal, but it also true that many would not agree that "general intelligence" is necessarily "human-like".
- AI should attempt to give a machine subjective conscious experience (consciousness or sentience). (This is what John Searle and most academic sources call "strong AI"). Even if it was clear how this could be done, it is an open question as to whether consciousness is necessary or sufficient for intelligent problem-solving.
- The question at issue is this: do any of these senses of "human like" represent the majority of mainstream AI research? Or do each of these represent the goals or methodology of a small minority of researchers or commentators? ---- CharlesGillingham (talk) 08:48, 7 October 2014 (UTC)
- @Felix: What do you mean by "human-like"? Is it any of the senses above? Is there are another way to construe it I have overlooked? I'm am still unclear as to what you mean by "human-like" and why you insist on including it in the lede. ---- CharlesGillingham (talk) 09:23, 7 October 2014 (UTC)
- One other meaning occurs to me now I have slept on it. The phrase "human-like" could be used as shorthand for "'human-like', whatever that means", i.e. it could be denoting a deliberately fuzzy notion that AI must clarify if it is to succeed. Mary Shelley galvanized Frankenstein's monster with electricity - animal magnetism - to achieve this end in what was essentially a philosophical essay on what it means to be human. Biologists soon learned that twitching the leg of a dead frog was not what they meant by life. People once wondered whether a sufficiently complex automaton could have "human-like" intelligence. Alan Turing suggested a test to apply but nowadays we don't think that is quite what we mean. In the days of my youth, playing chess was held up as an example of more human-like thinking - until the trick was pulled and then everybody said, "oh no, now we know how it's done that's not what I meant". Something like pulling inferences from fuzzy data took its place, only to be tossed in the "not what I meant" bucket by Google and its ilk. You get the idea. We won't know what "human-like" means until we have stopped saying "that's not what I meant" and started saying, "Yes, that's what I mean, you've done it." In this light we can understand that some AI researchers are desperate to make that clarification, while others believe it to be a secondary issue at best and prefer to focus on "intelligence" in its own right. — Cheers, Steelpillow (Talk) 09:28, 8 October 2014 (UTC)
I'm unhappy with it for another reason. "Artificial Intelligence" in computer science I think is now a technical term that is applied to a wide range of things. When someone writes a program to enable self driving cars - they call it artificial intelligence. See Self Driving Car: An Artificial Intelligence Approach "Artificial Intelligence also known as (AI) is the capability of a machine to function as if the machine has the capability to think like a human. In automotive industry, AI plays an important role in developing vehicle AI plays an important important role in developing developing vehicle vehicle technology." For a machine to function as if it had the capability to think like a human - that's very different from actually emulating human-like intelligence. Deep Blue was able to do that also - to chess onlookers - it acted as if it had the capability to think like a human at least in the limited realm of a chess game. In the case of the self driving car, or Deep Blue - you are not at all aiming to pass the Turing test or make a machine that is intelligent in the way a human is. Indeed, the goals to make a chess playing computer or a self driving car are compatible with a belief that human intelligence can't be programmed.
I actually think that way myself, persuaded by Roger Penrose's arguments - I think myself that no programmed computer will ever be able to understand mathematics in the way a mathematician does. Can never truly understand what is meant by "this statement is true" - just feign an understanding of truth, continually corrected by its programmers when it makes mistakes. His argument also extends to quantum computers and hardware neural nets. He doesn't think that hardware neural nets capture what the brain does, but that there is a lot going on within the cells which we don't know about that are also relevant as well as other forms of communications between cells.
But still, I accept that in tasks of limited scope such as chess playing or driving cars, they can come to out perform humans. This is nothing to do with weak AI or strong AI as I think both are impossible myself. Except perhaps with biological machines (slime moulds) or computers that in some way can do something essentially non computable (recognize mathematical truth) - if so they have to go beyond ordinary programming and go beyond ordinary quantum computers also to some new thing.
So - I know that's controversial - not trying to persuade you of my views - but philosophically it's a view that some people take, including Roger Penrose. And is a consistent view to have. Saying that the aim of AI is to create human like intelligence - that's making a philosophical statement that the things programmers are trying to achieve with self driving cars and with chess playing computers are on a continuum with human intelligence and we just need more of the same. But not everyone sees it that way. I think AI is very valuable, but not in that way, not the direction it is following at present anyway.
Plus also - the engineers of Google's self driving cars - are surely not involved in a "goal of emulating human-like intelligence" except in a very limited way.
Rather, their main aim is to create machines that are able to take over from humans in a flexible human environment without causing problems - and to do that by emulating human intelligence to whatever extent is necessary and useful to do that work.
Also in another sense, emulating human intelligence is too limited a goal. In the case of Deep Blue the aim was to be better at chess than any human - not to just emulate humans. Ideally also the Google self driving cars will be better at driving than humans. The aim is to create machines that in their own limited frame of reference are better than humans - and using designs inspired by capabilities of humans and drawn from things that humans can do. But not at all to emulate humans including all their limitations and faults and mistakes and accidents. I think myself very few AI researchers have that as a goal. So not sure how the lede should be written, but I am not at all happy with "goal of emulating human intelligence" - that is wrong in so many ways - except for some science fiction stories. I suggest also that we say "In Computer science" to start the second sentence, whatever it is, to distinguish it from "In fiction" where the goal often is to emulate human intelligence to a high degree as with Asimov's positronic robots. Robert Walker (talk) 00:28, 16 October 2014 (UTC)
- To limit the primary focus of AI to human-like intelligence, as User:FelixRosch originally sought to do with the lede, would be to ignore the successes of the field and focus on the aspect of the field that is always ten years in the future. Robert McClenon (talk) 21:50, 22 October 2014 (UTC)
Citations
One thing I did notice is that about half the page content is references and additional links of one kind or another - the article itself barely makes half way through the page. This seems absurd to me and I would suggest a severe rationalisation down to a few key secondary or tertiary works, with other sources (especially primary research) cited only where it is essential. — Cheers, Steelpillow (Talk) 20:32, 19 November 2014 (UTC)
- Please do not remove references from the article. The references record the sources used when writing it. This is a top-level article which needs references from many different sources addressing the fields mentioned. References to iconic original papers and lectures are an important part of the history of a subject. The references are grouped by field (the subtitles for the fields make the listing a bit longer than it otherwise would be but add value for the reader) and followed by an alphabetical list of citations, mainly for books and journals. These provide different views of the literature supporting the article contents, depending on what the reader needs.There are eleven citations which are not linked to from the references. We can probably remove those, which would shorten that list a bit. --Mirokado (talk) 09:46, 20 November 2014 (UTC)
- I would beg to differ. Your comment mixes two separate issues.
- WP:CITE is about verifying content, it is not about historical embellishment. By all means mention famous and pivotal papers, but those mentions should be supported by citing secondary and tertiary sources, not the papers themselves. Remember, the paper is being mentioned because of its importance, and the importance needs to be verified. No paper can verify its own importance. If the reader wants to dig deeper than the linked articles go, then the place to turn to in the first instance is the sources which do cite the primary papers: such lists will be a lot more complete than anything we can give in the present article. — Cheers, Steelpillow (Talk) 10:19, 20 November 2014 (UTC)
- If you really feel such a list can be justified in its own right (and I do not disagree there), can I suggest that since it is far the biggest part of this page it should be hived off to a List of primary sources in artificial intelligence? That way, it wouldn't clog up this article's citations. — Cheers, Steelpillow (Talk) 10:30, 20 November 2014 (UTC)
- I don't think that the length of the references is a problem that we need to solve. It's certainly not a problem for the reader, since they never scroll down that far. It's not a problem for Wikipedia, as it is WP:NOTPAPER. The complexity of these references is a bit of a problem for us, the editors of this article. Even this isn't a huge problem: per WP:CITE, it's always okay to add sources in any way that is complete and unambiguous, and eventually other editors (who enjoy this sort of thing) will bring them in line with the article's format.
- As I've said above, the most difficult problem in editing this article is finding the right weight (per WP:UNDUE), and the format of these references also provides proof that each topic belongs here. ---- CharlesGillingham (talk) 17:24, 20 November 2014 (UTC)
- @Steelpillow seems to recognize these concerns by making what appears to be a good solution in offering the editors to start of page for "List of primary sources in artificial intelligence". The list in its current form is not a comprehensive list, nor is it an exhaustive list, though it does take up about literally half of the article size which is un-needed here in the article itself. Linking to the moved material can be retained in the article, and the non-comprehensive and non-exhaustive list can be moved and can have its own page. Cheers. FelixRosch (TALK) 17:49, 20 November 2014 (UTC)
- Can you guys point me to an example of what you're talking about? And also, I'm very serious about the undue weight thing -- it matters; I've been editing the article for 7 years now and it comes up over and over. It's nice to be able show that every important point appears in most of the most reliable sources. ---- CharlesGillingham (talk) 22:45, 20 November 2014 (UTC)
- @Steelpillow; We're still with you on this. The material mostly appearing after the References section, such as "Other sources" should have its own page since it does not directly relate to the article itself. FelixRosch (TALK) 18:14, 21 November 2014 (UTC)
- @CharlesGillingham: There are many ways to cut the citation cake, so what I give below is a personal reaction. It expresses the meat of my complaint, though not necessarily the optimum solution.
- First off, all those bullet lists that cite Russell & Norvig and a bunch of others. Citing just Russell & Norvig would be fine, maybe one other if the factoid is particularly contentious. Looking at the sheer quantity of them in some paragraphs and the repetition in the list of Notes, I suspect that many of these could be reduced to a single citation at the end of the paragraph.
- If a source is used for many citations, use it for as many others as it is appropriate for. Even if a standard reference work gets cut from the citations altogether, that is no problem. Citing every standard reference work around is not the job of the main article content.
- Where a note cites a work given in full in one of the lists, say Russell & Norvig 2003, all such works should be collected alphabetically in a single appropriately-headed list so that the poor reader can actually find them. For example a Bibliography would be a good list.
- Works used to build the article but not actually cited should also be included in the bibliography.
- Sub-lists such as History of AI or AI textbooks are not appropriate in all that because they break the alphabetical listing, and the work may be cited in other sections than History anyway.
- Sub-lists are more appropriate for Further reading. These are books not plundered for the article but still of interest to the keen reader. If the above were done, the length of the Further reading section would then dictate its fate. If it were too long then it should form the basis of a standalone List of works on AI, and a copy of the bibliography should be merged into it.
- At the moment, it is actually quite hard to take any given example and follow the trail as to its relevance. And that's my point. Russell & Norvig only sprang up because of the relentless repetition. As the citations get refactored, it will become easier to pick out more examples.
- Does that help? — Cheers, Steelpillow (Talk) 20:08, 21 November 2014 (UTC)
- @Steelpillow; We're still with you on this. The material mostly appearing after the References section, such as "Other sources" should have its own page since it does not directly relate to the article itself. FelixRosch (TALK) 18:14, 21 November 2014 (UTC)
- Can you guys point me to an example of what you're talking about? And also, I'm very serious about the undue weight thing -- it matters; I've been editing the article for 7 years now and it comes up over and over. It's nice to be able show that every important point appears in most of the most reliable sources. ---- CharlesGillingham (talk) 22:45, 20 November 2014 (UTC)
- @Steelpillow seems to recognize these concerns by making what appears to be a good solution in offering the editors to start of page for "List of primary sources in artificial intelligence". The list in its current form is not a comprehensive list, nor is it an exhaustive list, though it does take up about literally half of the article size which is un-needed here in the article itself. Linking to the moved material can be retained in the article, and the non-comprehensive and non-exhaustive list can be moved and can have its own page. Cheers. FelixRosch (TALK) 17:49, 20 November 2014 (UTC)
- These are topically bundled (WP:CITEBUNDLE), short citations (WP:CITESHORT), using list defined reference (WP:LDR). All of these are accepted techniques in Wikipedia, although I agree it's rare to see so them used with such enthusiasm in one article.
- Each footnote is on a particular topic, so it makes no sense to combine them based on what paragraphs they happen to appear in. They are used in multiple paragraphs, and some paragraphs contain multiple topics.
- The "bibliography" of this article is called References (following standard practice in Wikipedia for WP:CITESHORT citations0. If you want to sort the textbooks and histories into the main list so that there is only one alphabetical list, that's fine. Note that you can click on a short citation and it takes you to the main citation.
- I have cited three or four sources for each major topic because this shows that the topic has enough weight to be included in the article -- if it's covered by most of the most reliable sources, then this article should cover it. Of course, weight is not something that concerns the reader, but I honestly don't think that readers use footnotes all that often. I'm not sure where else we could document this, except at Talk:Artificial intelligence/Textbook survey. If it really bothers you, we could cut these down to just R&N for some of these. I wouldn't cut down any of the footnotes that contain other information (such as the footnote 1, for example).
- I'm not sure what mean about it being "hard to take an example and follow the trail". You read a sentence, click on the footnote, and there's your relevance: you see that the topic is covered by most of the most reliable sources.
- Again, I want to point out that size of the reference does not harm the reader in any way, and is not a problem that needs to be solved -- there are other more urgent issues: the Applications and Industry section is basically unwritten and Fiction section is a travesty. ---- CharlesGillingham (talk) 02:20, 23 November 2014 (UTC)
I have started to tidy up the remaining errors reported by User:Ucucha/HarvErrors.js. These are for citations which specify a |ref=
parameter with no corresponding {{harvnb}}
or whatever reference.
- For the textbooks and history subsections, which list general background sources some of which are specifically referenced, it seems better to retain the citations but remove the parameter – it is easy to restore the parameter if a reference is added.
- For the others, I will at least mostly remove the citations and list them here for any further consideration.
I will continue to tweak for source consistency as I go, with the parameter order for citations roughly: last etc, year/date, title, publisher etc, isbn etc, url, ref. Having last, first, date in that order helps when matching citation to callout and having last first (!) helps when checking the citation sorting. --Mirokado (talk) 16:42, 6 December 2014 (UTC)
Removed:
- Dreyfus, Hubert (1979). What Computers Still Can't Do. New York: MIT Press. ISBN 0-262-04134-0.
- Forster, Dion (2006). "Self validating consciousness in strong artificial intelligence: An African theological contribution" (PDF). Pretoria: University of South Africa.
- Lakoff, George (1987). Women, Fire, and Dangerous Things: What Categories Reveal About the Mind. University of Chicago Press. ISBN 0-226-46804-6.
- Moravec, Hans (1976). "The Role of Raw Power in Intelligence". Retrieved 30 August 2007.
- Newell, Allen; Simon, H. A. (1963). "GPS: A Program that Simulates Human Thought". In Feigenbaum, E.A.; Feldman, J. (eds.). Computers and Thought. New York: McGraw-Hill.
- Serenko, Alexander; Detlor, Brian (2004). "Intelligent agents as innovations" (PDF). AI and Society. 18 (4): 364–381. doi:10.1007/s00146-004-0310-5.
- Serenko, Alexander; Ruhi, Umar; Cocosila, Mihail (2007). "Unplanned effects of intelligent agents on Internet use: Social Informatics approach" (PDF). AI and Society. 21 (1–2): 141–166. doi:10.1007/s00146-006-0051-8.
--Mirokado (talk) 17:02, 6 December 2014 (UTC)
removed conceptual foundations section
Sorry for my boldness, but I think this section should be more readable. Since it seems to my cursory glances that some, if not most, of these bullet points are included in the history section, I'm removing conceptual foundations section to the talk page. Xaxafrad (talk) 07:55, 23 November 2014 (UTC)
- I've edited this list and removed the items which are mentioned in the history section. Xaxafrad (talk) 08:52, 23 November 2014 (UTC)
Conceptual foundations
The conceptual foundations defining artificial intelligence in the second decade of the 21st-century are effectively best summarized as a list of strongly endorsed research pairings of contemporary 21st-century research topics as follows:
- Symbolic AI versus neural nets
- Reasoning versus perception
- Reasoning versus knowledge
- Representationalism versus non-representationalism
- Brains-in-vats versus embodied AI
- Narrow AI versus human-level intelligence. [1]
Several key moments in the history of AI have contributed to defining the major 21st-century research areas in AI. These early historical research areas from the last century, although by now well-rehearsed, are revisited occasionally with some recurrent reference to:
- McCullock and Pitts early research in schematizing digital circuitry
- Samuel's early checker player
From these followed the 4 more historical research areas currently being pursued in updated form which include
- means-ends problem solvers (Newell 1959)
- Nautral language processing (Winograd 1972)
- knowledge engineering (Lindsay 1980)
Major recent accomplishments in AI defining future research paths in the 21st-century have included the development of
- Solution to the Robbins conjecture
- Killer App ( and Gaming applications as a major force of research and innovation). [2]
The current major leading 21st-century research areas appear to be
- Knowledge maps
- Heuristic search
- Planning
- Machine vision
- Machine learning
- Natural language
- Software agents
- Intelligent tutoring
The most recent 21st-century trends appear to be represented by the fields of
- Soft computing
- Agent based AI
- Cognitive computing
- AI and cognitive science. [3]
- (end of removed section) So there it is...Sorry for stepping on anyone's toes. I'd be happy to help edit these
3625 list items into the history section if it's needed. Xaxafrad (talk) 08:29, 23 November 2014 (UTC)
Incorporating Franklin
I think we can incorporate Franklin as one of our sources. I went through bullet lists above and identified where in the article we cover the same topics. There's only a few question marks --- someone needs to read Franklin carefully and make sure that I am right so far. ---- CharlesGillingham (talk) 21:37, 30 November 2014 (UTC)
- Symbolic AI versus neural nets
- Under Approaches. A difference between Symbolic AI and (the earliest form of) Computational Intelligence and soft computing. Covered in detail in History of AI#The revival of connectionism.
- Reasoning versus perception
- Not sure; possibly under Approaches,relevant to the difference between Symbolic AI and Embodied AI. Or is Franklin talking about David Marr vs. Symbolic AI? Should we even mention Marr? This was not a particularly influential dispute, but Marr is covered in History of AI#The importance of having a body: Nouvelle AI and embodied reason.
- Reasoning versus knowledge
- Under Approaches, the difference between Knowledge based AI and the rest of Symbolic AI.
- Representationalism versus non-representationalism
- Under Approaches, the difference between Symbolic and Sub-symbolic AI.
- Brains-in-vats versus embodied AI
- Under Approaches,the difference between Embodied AI and Symbolic AI.
- Narrow AI versus human-level intelligence.
- Under Goals, relevant to the difference between general intelligence and all other goals.
Several key moments in the history of AI have contributed to defining the major 21st-century research areas in AI. These early historical research areas from the last century, although by now well-rehearsed, are revisited occasionally with some recurrent reference to:
- McCullock and Pitts early research in schematizing digital circuitry
- They are named in a footnote 24: "AI's immediate precursors", and in the first sentence of Neural networks. Covered in more detail in History of AI#Cybernetics and early neural networks.
- Samuel's early checker player
- Added this to "Golden years" sentence and footnote in History. Covered in more detail in History of AI#Game AI.
- means-ends problem solvers (Newell 1959)
- The General Problem Solver is not covered in AI, but this is covered in detail in History of AI#Reasoning as Search
- Nautral language processing (Winograd 1972)
- SHRDLU is mentioned in "Golden years" sentence and footnote in History
- knowledge engineering (Lindsay 1980)
- This is probably covered in Approaches under Knowledge-Based, or in the paragraph of History on expert systems (and History of AI#The rise of expert systems). I am not familiar with Lindsay 1980; could someone read Franklin and see what this is?
Major recent accomplishments in AI defining future research paths in the 21st-century have included the development of
- Solution to the Robbins conjecture
- Should be added to the last paragraph of History.
- Killer App ( and Gaming applications as a major force of research and innovation).
- Don't know what Franklin is getting at here
The current major leading 21st-century research areas appear to be
- Covered under Knowledge Representation. Or does he just mean "knowledge representation"? Also, why does he leave out "reasoning"?
- Planning
- Covered in Planning.
- Machine vision
- Covered under Perception.
- Machine learning
- Covered in Machine learning.
- Natural language
- Covered in Natural language processing.
- Software agents
- Possibly covered under Approaches in Intelligent agent paradigm, unless Franklin is saying something else ... to be checked.
- Intelligent tutoring
- We don't have this; I think it belongs under Applications.
The most recent 21st-century trends appear to be represented by the fields of
- Soft computing
- Added this under Approaches
- Agent based AI
- Probably covered under Approaches in Intelligent agent paradigm
- Cognitive computing
- Not sure what he means by this ... is this the hardware thing that IBM calls cognitive computing? The Wikipedia article on this term is horrific; there are only sources for IBM's hardware thing -- no sources at all for the more general term. Perhaps Franklin could be used to straighten out that article.
- AI and cognitive science.
- Similarly, not too sure about this either; cognitive science is relevant in many places in the article; not sure what trends he's talking about exactly.
- ---- CharlesGillingham (talk) 21:37, 30 November 2014 (UTC)
- ^ Franklin (2014), Cambridge Univ Press, pp15-16. The Cambridge Handbook of Artificial Intelligence (2014). [8]
- ^ Franklin, Cambridge Univ Press, pp22-24. The Cambridge Handbook of Artificial Intelligence (2014). [9]
- ^ Franklin, Cambridge Univ Press, pp24-30. The Cambridge Handbook of Artificial Intelligence (2014). [10]