Best way to teach humans some opening reprtoire

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Uri Blass
Posts: 8921
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Best way to teach humans some opening reprtoire

Post by Uri Blass » Sat Oct 24, 2020 9:17 am

You have a set of 1000 moves that you need to memorize in the opening in different positions.
You want to learn as much as possible in an hour by spaced repetition.

I wonder if there is a research what is the best way to do it when the instructions what to do is by a chess program and if it is possible to make a competition of programs to teach humans.

For competition between program A and program B the idea is that you take 200 humans and divide them randomly to 2 sets of 100 humans.
You give 100 random humans to learn for an hour by program A and 100 random humans to learn by program B for an hour and compare the results

After an hour of learning the program can decide how many positions you could memorize the best move and you get a test by some testing program that give you the problems by random order.
For example
If program A say that you memorized the first 40 positions then you need to get the first 40 positions in the reperoire in random order and if you get all the moves correctly in less than 3 seconds your score is 40 otherwise your score is significantly lower by some formula(for example you can decide that if you got x moves correctly in less than 3 seconds for x<40 the score is
40*((x/40)^4) or a different function.

You calculate the average score of humans who use program A and calculate the average score of humans who use program B to learn and the winner is the program with a better average score.(hopefully with a lot of humans luck of having better team of humans is not a significant factor).

A possible plan that is probably not the best is to repeat 1 until you make the right move in less than 3 seconds later repeat 1,2 in random order until you make the right moves in both of them in less than 3 seconds and then repeat 1,2,3 in random order until you get the right move in less than 3 seconds and continue in this way.

Not the best because I guess that it is better to reduce the number of times you repeat a position if you always make the right move in it and increase the number of times you repeat positions that you fail to memorize the right move often.

Ferdy
Posts: 4463
Joined: Sun Aug 10, 2008 1:15 pm
Location: Philippines

Re: Best way to teach humans some opening reprtoire

Post by Ferdy » Sat Oct 24, 2020 5:11 pm

I think chessable is doing spaced repetition. Learning would be better if during learning humans would give feedback to the positions, i.e. move is easy, normal, or difficult to find. This way the program would train more often on those positions that are difficult. A human feedback can be automatically taken by measuring how long does one take to find the solution move in the position.

Frank Quisinsky
Posts: 5400
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Full name: Frank Quisinsky
Contact:

Re: Best way to teach humans some opening reprtoire

Post by Frank Quisinsky » Sat Oct 24, 2020 5:38 pm

Hi Ferdy,

that's is right but to many of opening books by strong chess players are available.
The market confuse me.

Often older books have the better / readable comments for the own understandings!
Often not free of mistakes / blunders, but such things will make an opening book more interesting because the players try to find out.

The reason for me that I don't have many opening books in my chess book collection.
More interesting are biographys.

In my opinion the opening books by Quality chess have best material inside.
Great are the Sahovski overview GM Alexander Matanovic (Encyclopaedia of Chess Openings) or the work by IM John Watson from Gambit (4 opening books). Grandmaster Opening Preparation by Jaan Ehlvest from Quality chess is good for the start, much other books are good for a start to learn more.

The problem with Encyclopaedia of Chess Openings or other good overviews are the symbols. I like more if I can read a comment by a grandmaster. That's what I mean, really good learning is better with strong biography material or books with pure learning. Most are thinking I read 10 of the books and my chess will be better. Such books are very rarely but in my opinion we have it.

A good way to train openings is the BookUp software or to create an opening book for a chess engine by hand in self work ... step by step. Here I learn a lot but most with good chess books. Important here an engine with a good king safty with many pieces on board. I am using all the time Wasp for opening analyzes because I like that Wasp try to find out allways the aggressive pawn moves. King safty is very strong for the Elo Wasp have.

I like chess books a lot and have my own library.
For some days I added pics in the "About me" selection on my website.

A good chess book and a DGT board with Wasp as analyze engine ... Frank is happy.
Sorry that I speaking again from Wasp but it have a reason that i like the program so much (the first playing phase, openings and the earlier middle-game is most interesting in chess for myself).

Best
Frank
I like computer chess!

Ferdy
Posts: 4463
Joined: Sun Aug 10, 2008 1:15 pm
Location: Philippines

Re: Best way to teach humans some opening reprtoire

Post by Ferdy » Sat Oct 24, 2020 8:03 pm

Bookup is good indeed. You open your old books then input in bookup then analyze and backsolve. :D

For tabs and phones this one is good. There is training with random and sequential, speed test and other options like which positions to train (low scores, untrained, schedule).

Post Reply