TalkChess.com

BrendanJNorman wrote: ↑Mon Mar 23, 2020 11:29 pm It sounds like he wants the engine to be sentient, which as Dan said, will not be possible for a LOOOOONG time.

BrendanJNorman wrote: ↑Mon Mar 23, 2020 11:08 pm

Ovyron wrote: ↑Mon Mar 23, 2020 11:01 pm

BrendanJNorman wrote: ↑Mon Mar 23, 2020 10:52 pm I'll be impressed (legitimately) even if you even got up to 1800, but the way you're behaving is impossible to believe.

What if it was on rapid? Suddenly only blitz ratings matter?

Ovyron wrote: ↑Mon Mar 23, 2020 10:22 pm ...just yesterday I reached a winning position against a 2100 guy WITH EASE on a 5 0 game...

Ovyron wrote: ↑Tue Mar 24, 2020 1:51 am Yeah, that was a rare occurrence (by easy I mean that it was easy to do it for that game, not that I can do it easily always). My Rapid rating is 150 elo higher than my Blitz rating.

mclane wrote: ↑Mon Mar 23, 2020 9:43 pm
LC0 is the evidence for me that it is possible to build strategy over tactics and still play decent human chess.

There are two types of engines I would like to see more of.

1. A wild and crazy attacking style, closely emulating a Tal or Morphy like player, with a high contempt setting and swindle capability

or

2. A knowledge-laden program geared towards strategic/positional chess, strong in the endgame, but weaker tactically and still having the contempt setting and swindle feature, as above


I would actually be interested in rather spending money on such efforts than on the next super-engine (seriously!). Think about it, we're "swimming" in strength, but we're short on style. In my other post in this thread, I gave an example where a whopping 400 rating points could be sacrificed to achieve a credible "Tal"-like personality for most players.

My focus would be on creating a planning-based human-friendly program, where actual maneuvers and plans would be carried out, even if they could be refuted by strong human play.
...
I would pack as much knowledge into the engine as possible without fear of 'slowing' it down. Any improvements would be assessed based on its play and results against human competition. Tactical strength would be of secondary importance, but not altogether neglected. The burden of proof would be on the humans to refute the engine's plans.
---
The kind of program I envision could be a great sparring partner for humans, simulating other humans better than it can be done at the moment. Play against such an opponent would be a truly fascinating experience. As it becomes stronger, it could also become a great analysis tool.

Traditionally, the engines' strength is built primarily on tactics, which comes from search, so to emulate a human player we must have the priorities reversed and have search be crippled and take a back seat to evaluation based on chess knowledge.

I've always wanted to see an engine that could simply play stubbornly, attempting to carry out plans even if they didn't work. Normally, the inherent tactical strength of normal engines does not allow this. So, I think a realistic human-emulator must be crafted differently from the outset, with an emphasis on evaluation and chess moves generated based on what *looks* good in view of it, rather than on what works tactically based on deep calculation.

Don't most humans make moves because they just "look right" to them, and not necessarily because they're absolutely sound? Likewise, a human-emulator should play in a similar fashion. Each move should have a knowledge-based purpose first, with tactical soundness a secondary factor, and be heavily weighted towards the former.

mclane wrote: ↑Mon Mar 23, 2020 9:43 pmLC0 is the evidence for me that it is possible to build strategy over tactics and still play decent human chess.

There has been far less progress on the pure human-emulator front, with Dietrich Kappe's Badgyal neural nets coming the closest until now.

Dann Corbit wrote: ↑Tue Mar 24, 2020 9:28 am

There has been far less progress on the pure human-emulator front, with Dietrich Kappe's Badgyal neural nets coming the closest until now.

I am curious to know by what measure it is known that the programming of Dietrich Kappe is closer than other efforts.

Wat exactly is a Badgyal neural net? I have never heard of it.

Ras wrote: ↑Tue Mar 24, 2020 7:49 am Artificial neural networks have as little to do with human intelligence as using a compressor to press manure through a hole in the wall has to do with human digestion.

zullil wrote: ↑Mon Mar 23, 2020 11:44 pm

BrendanJNorman wrote: ↑Mon Mar 23, 2020 11:29 pm

jp wrote: ↑Mon Mar 23, 2020 11:11 pm But clearly mclane does not just want really good chess. He wants something else, or something more.

It sounds like he wants the engine to be sentient, which as Dan said, will not be possible for a LOOOOONG time.

If an engine were sentient, and wanted to play chess well, it would likely install Stockfish and Lc0.

carldaman wrote: ↑Tue Mar 24, 2020 5:43 am I for one have always been interested in a chess program that can emulate human play, including some of the human weaknesses.