|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
| Jul-08-16 | | john barleycorn: <Sally Simpson: Hi John,
I thought I had it right, the problem is the actual game and the computer has to solve it. But they are not playing a game. It has no will to win, it fears not losing. It just calculates. ...> It uses an algorithm and picks the best *solution* according to his evaluation function. And it just *hopes* this produces a win.
Which today it does constantly against 98% of players. <Fears not losing> yes, that is a strength of these machines. I have never seen any of the things getting red ears and cheeks when their king was under attack. |
 |
Jul-08-16
 | | Sally Simpson: Hi John,
I don't think it hopes. They usually have it all worked out. Hope chess is a human trait. Another lad mentioned: "Houdini sacrifices 3 pawns for initiative." A computer does not know what the initiative is. That is us putting human thinking into the play of a machine. HeMateMe thinks Tal would love this era.
Then we recall the Tal quote.
"You must take your opponent into a deep dark forest where 2+2=5, and the path leading out is only wide enough for one." 2+2=5 will never compute.
There are no deep dark forests when machines meet. There is no landscape, no drama, no twists in the tail. No fun. |
 |
Jul-09-16
 | | OhioChessFan: <Sally: Humans blunder and win games. a human will knowingly play not the best move to set OTB problems for their opponent to solve. This is something a machine could never do. It has to play what it considers is the best move all the time and every time.
>
Talking programming is beyond my pay grade, but let me take a shot at this. I think it'd be rather simple to program a little subroutine to play a sub-optimal move based on the resulting position. Something like: If top 10 moves have positive evals of <.05, then play 11th best move if opponent has >10 only moves in next 30 plies. |
 |
Jul-09-16
 | | alexmagnus: Or take a similar approach as AlphaGo did in Go. Train a neural network on human games. And them combine the NN evaluations wih the evaluations of a comventional engine. Then you get an engine that plays better than humans but "human". |
 |
Jul-09-16
 | | alexmagnus: <A computer does not know what the initiative is.> Doesn't it? After all, somehow it <does> evaluate positions with imbalanced material correctly, which means it does understand conceptions like positional advantage, and yes, initiative. It may not know the word "initiative" but you do not have to have a word for something to know that this something exists. |
 |
Jul-09-16
 | | HeMateMe: White Zombie
More Human Than Human
<https://www.youtube.com/watch?v=E0E...> |
 |
Jul-09-16
 | | Sally Simpson: Hi Alex,
I was just commentating on how we use human terms in a wrong context when looking at a computer game. I suppose it's inescapable as we make attempts to understand and explain what is going on. |
 |
Jul-09-16
 | | morfishine: Who knows whats best...maybe we are left with 'Father Knows Best' and thats it ***** |
 |
Jul-09-16
 | | AylerKupp: <morfishine> Perhaps someone will be sufficiently motivated to write a killer chess program and name it 'Father'. |
 |
| Jul-09-16 | | scholes: <Sally> Check out this game Stockfish vs Jonny, 2014
Stockfish sacrifices a full rook at move 22. Then it doess nothing to regain the material till move 40. When it sacrifices another rook to win back the investment. It was voted by chess.com as crazy game of the year. |
 |
Jul-09-16
 | | RandomVisitor: <scholes>In that game, after 21...a5 click for larger viewKomodo-10-64bit:
+0.30/40 22.Nac4 Kg8 23.Qa3 f6 24.Nf3 Nd3 25.Rfc1 Nxc1 26.Rxc1 Nb4 27.Ne1 g6 28.Nxc2 Nxc2 29.Rxc2 Kg7 30.g3 Rhd8 31.h3 Rd7 32.Ra2 Ra8 33.Nb2 Qc1+ 34.Kh2 Rb7 35.Nd3 Qxa3 36.Rxa3 Rb5 37.Nc5 Kf7 38.Kg2 Ke7 39.Ra4 Rb4 40.Ra1 Kd6 41.Kf3 Kd5 42.Kf4 h6 43.Ra2 Rbb8 44.h4 Rb4 45.g4 g5+ 46.hxg5 fxg5+ 47.Kg3 Rbb8 <+0.21/40 22.Nb5 c1Q 23.Raxc1 Qxc1> 24.h4 Qc2 25.Nd6 g6 26.Ndxf7 Kg7 27.Nxh8 Kxh8 28.Qxa5 Kg7 29.Qb5 Rc7 30.Nc4 Qd3 31.h5 Re7 32.Qb8 Qf5 33.hxg6 hxg6 34.Ra1 Qc2 35.Qg3 Nf6 36.Nd6 Nc6 37.Qg5 Rd7 38.Qc5 Qb2 39.Rf1 Qxb3 40.e4 Nd8 41.e5 Nd5 42.Qc8 Nb6 43.Qc1 Ne6 44.Qe3 Qxe3 45.fxe3 Nd5 46.Ne8+ Kh6 47.Nf6 Rd8 48.Kf2 Rf8 |
 |
| Jul-10-16 | | falso contacto: Black King was miserable all the way tho. |
 |
Jul-10-16
 | | Sally Simpson: Hi Scholes,
Thanks for the link, seen it before, still as unimpressed now as I was then. I see they made it GOTD. What's next? 'Sportsman of the Year.' Here:
Substitute 'song. for 'game.'
https://www.youtube.com/watch?v=Gx_... |
 |
Jul-11-16
 | | alexmagnus: As for what a computer knows or not. It seems to me many people mix up three notions: knowledge, intelligence and consciousness. Even the most primitive computer possesses knowledge. Actually anything connected with some kind of explicit or implicit database does. Just as easy is the answer whether a computer is conscious - a clear "no". With intelligence, there are debates. I'm on the "yes" side, because most of the "no" arguments IMO mix it up with consciousness. Intelligence is the ability to perform some tasks for which we humans require thinking. It is not necessarily being aware of that ability (after all, already the word "aware" suggests consciousness). In this sense, computers are intelligent, though their intelligence is specific, not general (IBM's Watson may be the first step towards general intelligence). The pyramid is this:
Knowledge < specific intelligence < general intelligence < consciousness. Each "bigger" part cannot exist if the smaller part is absent. Chess computers are on the specific intelligence level. Modern computer technology generally somewhere in between specific and general intelligence. And it will take some time for comps to gain consciousness. Nobody still knows what consciousness is. There is that somewhat disturbing theory that consciousness emerges spontaneuosly in any intelligent system that is complex enough, with the degree of the necessary complexity being yet unknown. Disturbing because if this is the case, one day we may create a conscious computer without having aimed for it. |
 |
Jul-11-16
 | | Schwartz: I have a cat, and I'm convinced mice have consciousness.. also they can make about 200 sounds. Anyway, humans have access to a really pretty good assortment of tools to operate. Some computers should have a vibrant 'ecosystem' to support their purposes.. but they're(the computers) not very similar to humans.. and I don't see them having consciousness for the next 1000 years. |
 |
| Jul-11-16 | | Everett: We don't even understand how consciousness works, how it's created in living things, how it operates... Yet, we feel we can create it in a computer. Remarkable. |
 |
Jul-11-16
 | | Appaz: <Everett> We should be grateful that some people don't think this way. The phenomena of flying was not fully explained, but that didn't stop some folks from building more or less successfully flying constructions. Experimentally mimicking the nature is often a good way to reach an explanation. |
 |
| Jul-15-16 | | posoo: now DIS man dos not play many DRAUGHS. i have many respect for HIM~! |
 |
| Jul-18-16 | | Everett: < Appaz: <Everett> We should be grateful that some people don't think this way> Maybe you should keep your shoulds to yourself.
And to compare consciousness to flying to prove a point is quite an obtuse idea. Perhaps you should choose another 😀 |
 |
Jul-18-16
 | | Appaz: <<Everett> Maybe you should keep your shoulds to yourself.> Weird expectations for a public forum.
<And to compare consciousness to flying to prove a point is quite an obtuse idea. Perhaps you should choose another> No, I'm quite happy with that one. |
 |
Aug-03-16
 | | Penguincw: Komodo 10 wins the TCEC Season 9, Stage 2 Championship, with a score of 22.5/30, tied with Stockfish 110616 (edged it out on tiebreaks, 320.25-314.00). Komodo was also the top seed here, with a rating of 3228, 6 points ahead of Stockfish. Komodo also gained 2 rating points for its effort...
http://tcec.chessdom.com/archive.php |
 |
Aug-13-16
 | | Ron: Rating list for chess engines:
http://www.computerchess.org.uk/ccr... |
 |
| Sep-14-17 | | scholes: Komodo wins knight odds game against Nakamura
http://talkchess.com/forum/viewtopi... |
 |
| Nov-01-17 | | SChesshevsky: A couple of quick questions:
In competitions like TCEC's it seems like the machines use book for the first half dozen or so moves. Does the computer pick which book opening is played? If not, how is it decided? Given the advance in machine learning, I'd guess that a high powered computer with an adequate database could narrow down the entire chess opening spectrum and distill the best array of lines that could gain an advantage for white and conversely always or just about equalize for black. This computer program would not be interested in anything else but finding those best "sure fire" openings. With criteria, say, position after move 16. or so. Has anyone heard of any development efforts for such a program? Thanks! |
 |
| Nov-01-17 | | markz: <SChesshevsky: Given the advance in machine learning, I'd guess that a high powered computer with an adequate database could narrow down the entire chess opening spectrum and distill the best array of lines> In my opinion, it is almost impossible, because the number of possible openings is almost unlimited. For machine learning, we need to know the best moves so that machine can learn them, which is almost impossible also. I think it is possible that the best chess engines fight with each other million and million times, they will learn which moves are the better moves from their match results, and then become better and better. |
 |
 |
 |
|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |