What can I say - that this is the answer one can only wish. Bravo!
The information about this KBBKNN situation I've read around 1987, must have been a little deformed by that magazine. I've took them too seriously.
Now, I am going to investigate another piece I recall and I couldn't find it online until now. This time from the Science magazine sometimes during 1980's. The title I remember was "Never Out of Sorts".
The tables of contents for Science magazine are online. Looking through these might jog your memory. But there are quite a lot of issues.
Recently in another topic I mentioned the "two bishops against two knights" chess endgame problem. I claimed it was investigated over two decades ago by a computer program and established that it is a win situation for the two bishops' side. Then I was unable to Google a solid reference for my claim.
I believe that subject to the ambiguity in what is meant by "a win situation for the two bishops", your recollection is correct.
The 6-piece pawnless endgames were were first analyzed systematically by Lewis Stiller starting in the late 1980s and reported in his papers in 1991 and 1992. The storage technologies available at this time meant that only summarized results could be saved, such as the longest win, and the total number of wins, draws and losses. I can't find these papers online, but the results also appear in Stiller (1995) and there's a summary of the state of the art in Thompson (1996).
For the KBBKNN ending Stiller only analyzed positions with the two bishops on opposite coloured squares (and I think with white to move), and reported that the longest win for white was 37 moves and the percentage of wins for white was 63%.You probably also want to note Stiller's caveat:
The percent-win can be misleading because of the advantage of the first move in a random position—White can often capture a piece in one move—and because it includes positions in which Black is in check.
So I think if you said "mostly a win for the two bishops from a random position with bishops on opposite-coloured squares, with the player with the bishops to move" that would be a fair summary of the facts.
Modern tablebases usually also include positions with the two bishops on the same colour square, so that analyses of these databases will give different results to Stiller. For example, according to Kirill Kryukov, the KBBKNN positions split like this:
With white (bishops) to move: 28429 losses, 885809752 draws (76%), 282912378 wins (24%)
With black (knights) to move: 54327970 losses (4%), 1247006005 draws (96%), 154105 wins
How could you have found this using Google? Well, it always helps to know of specialized databases to search (because the results tend to be of higher quality). I used Google Scholar to search for academic papers relevant to the keywords "6-piece chess endgame" and that returned Thompson (1996) as the first hit, and reading Thompson's summary of the state of the art led me to the Stiller papers. Of course, domain expertise is a big help too: I realised after discovering Stiller (1995) in the course of this search that I have a copy of this on my bookshelves.
References
Fairly recent, as far as I know; probably no earlier than the 1980s. (This is just a guess based on vague memory.) I'm not even sure it has spread much beyond people whose interest in Latin is specifically linguistic. (For instance I don't know that there are any pedagogical materials -- as opposed to linguistic treatises -- that use this spelling, though there might be.)
Spelling Latin with u has always been there (but as a tiny minority of texts). Here are some occurrences of omnia uincit amor over the years: 1603, 1743, 1894, 1974.
If you compare the frequencies of vincit and uincit on Google Ngram viewer, you'll see that the u spelling has always been present at a low frequency. There doesn't seem to be any noticeable recent trend (other than the general decline of Latin as a proportion of printed material). I tried a few other Latin words and got similar results.
A statement like this is attributed to Schoenberg by a number of people, but I can't find a specific reference either. Perhaps it was just something he said orally, without ever writing it anywhere.
The earliest reference I can track down is from 1952. In Roger Sessions: a biography (2008), Andrea Olmstead writes:
[In 1952] Sessions published "Some notes on Schoenberg and the 'method of composing with twelve tones'." At the head of the article he quoted from one of Schoenberg's letters to him: "A Chinese philosopher speaks, of course, Chinese; the question is, what does he say?" Sessions [had performed] the role of a Chinese philosopher in Cleveland.
(The work that Sessions had performed this role in appears to have been Man who ate the popermack in the mid-1920s.)
Sessions' essay (originally published in The Score and then collected in Roger Sessions on Music) begins:
Arnold Schönberg sometimes said 'A Chinese philosopher speaks, of course, Chinese; the question is, what does he say?' The application of this to Schönberg's music is quite clear. The notoriety which has, for decades, surrounded what he persisted in calling his 'method of composing with twelve tones', has not only obscured his real significance, but, by focusing attention on the means rather than on the music itself, has often seemed a barrier impeding a direct approach to the latter.
An entertaining later reference to this quotation appears in Dialogues and a diary by Igor Stravinsky and Robert Craft (1963), where Stravinsky tabulates the differences between himself and Schoenberg, culminating in this comparison:
Stravinsky: ‘What the Chinese philosopher says cannot be separated from the fact that he says it in Chinese.’ (Preoccupation with manner and style.)
Schoenberg: ‘A Chinese philosopher speaks Chinese, but what does he say?’ (‘What is style?’)
It's hardly fair to call EY Egan's 'biggest fan', but this is nonetheless amusing. The actual disrecommendation for the book was a little hard to find buried in all the clever analysis:
the novel never delivers the emotional impact that it promises. Ned Beauman in SFX calls it a “tepid meditation on fatherhood and Middle Eastern democracy,” which is a fair summary. Egan’s characterisation is simply not good enough to support the story he wants to tell.
I don't see how those novels could have been an inspiration? I've read them when I was just awakening (~2005) and even then I noticed the sharp absence of any artificial intelligence. I believe Greg Egan's idea of the future is still a serious possibility. After all, as with aliens, the only example of something resembling generally intelligent, aware and goal-oriented agents are we ourselves.
If there was an inspiration then I would suspect others to be a more likely source.
I haven't read the book, but it looks rather like that he portrays this movement as a conspiracy to live off the money of nonconformists that is hidden under a massive amount of writings about rationality and pillowed by the little cherry on the cake that is AI going FOOM (rapture of the nerds).
I don't see how those novels could have been an inspiration?
Yudkowsky describes Egan's work as an important influence in Creating Friendly AI, where he comments that a quote from Diaspora "affected my entire train of thought about the Singularity".
To reiterate, people do not need to know or understand the Bayesian rules of forgetting in order to successfully solve this problem. Nobody used this approach to solving the problem - as far as I am aware - but the vast majority obtained the correct answer nontheless. Correct reasoning is given on http://en.wikipedia.org/wiki/Sleeping_Beauty_problem - and in dozens of prior comments on the subject.
The Wikipedia page explains how a frequentist can get the answer ⅓, but it doesn't explain how a Bayesian can get that answer. That's what's missing.
I'm still hoping for a reference for "the Bayesian rules of forgetting". If these rules exist, then we can check to see if they give the answer ⅓ in the Sleeping Beauty case. That would go a long way to convincing a naive Bayesian.
The Sleeping Beauty Challenge
Maybe I'm naive, but I actually think that we can come close to consensus on the solution to this problem. This is a community of high IQ, aspiring rationalists.
I think it would be a good exercise to use what we know about rationality, evidence, biases, etc. and work this out.
I propose the following:
I will write up my best arguments in favor of the 1/2 solution. I'll keep it shorter than my original post.
Someone representing the thirders will write up their best arguments in favor of the 1/3 solution
Before reading the others' arguments, we will assume that they are right, and that reading it will only confirm our beliefs (this is hard to do, but I find that this approach can be helpful)
We cannot respond for at least 24 hours. (this will give us time to digest the arguments, without just reacting immediately)
We will then check to see if there is agreement
If we still disagree, we can have some discussion (say, via email) to see if progress can be made
We will post our original two arguments and conclusion here (maybe in a new post)?
What do you think?
I tried to set this up in such a way to reduce some of the known biases that prevent agreement. Am I missing something?
Possible pitfall: if we come to an agreement, people who disagree with our conclusion might say it's because one of us was a poor representative of their viewpoint. However, I think we'd still move a step towards consensus.
What say you?
I think this kind of proposal isn't going to work unless people understand why they disagree.
Well, there is not anything wrong with Bayes' Law. It doesn't model forgetting - but it doesn't pretend to. I would not say you have to "abandon" Bayes' Law to solve the problem. It is just that the problem includes a process (namely: forgetting) that Bayes' Law makes no attempt to model in the first place. Bayes' Law works just fine for elements of the problem involving updating based on evidence. What you have to do is not abuse Bayes' Law - by using it in circumstances for which it was never intended and is not appropriate.
Your opinion that I am under some kind of obligation to provide a lecture on the little-known topic of Bayesian forgetting has been duly noted. Fortunately, people don't need to know or understand the Bayesian rules of forgetting in order to successfully solve this problem - but it would certainly help if they avoid applying the Bayes update rule while completely ignoring the whole issue of the effect of drug-induced amnesia - much as Bradley Monton explains.
You're not obliged to give a lecture. A reference would be ideal.
Appealing to "forgetting" only gives an argument that our reasoning methods are incomplete: it doesn't argue against ½ or in favour of ⅓. We need to see the rules and the calculation to decide if it settles the matter.
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
= 27d567bc2d48d9f40e9ccc20ea370b15)
= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Front page: missing author
The front page for Facing the Singularity needs at the very least to name the author. When you write, "my attempt to answer these questions", a reader may well ask, "who are you? and why should I pay attention to your answer?" There ought to be a brief summary here: we shouldn't have to scroll down to the bottom and click on "About" to discover who you are.