Planning

From Chessprogramming wiki
Jump to: navigation, search

Home * Knowledge * Planning

Plan is law, fulfillment is duty, over-fulfillment is honor! [1]

Planning,
the process of applying knowledge, thinking and prediction to create and maintain a plan, a temporal set of intended actions to achieve a certain goal, in chess, to finally win the game.

Terms of Planning

In 1997, Jan van Reek [2] categorized planning in chess to three main characteristics, tactics, positional play, and strategy, where positional play serves as guideline for planning of a strategy [3]:

Terms of Planning Aim Dominating abstraction
Tactics Short To gain material or to mate Matter
Positional play Medium To improve the position Space
Strategy Long To swing the balance Time

Long-range Planning

Long-range planning by human chess players is usually made on the basis of reasoning, experience, and intuition, as investigated by De Groot in 1946 [4].

Search and Evaluation

Conventional depth-first alpha-beta searchers have no real sense of planning other than minimaxing their score inside their search horizons. Inside chess programs, planning is often implicit due to look ahead and certain evaluation features and their associated weights. For instance, a bonus for a Rook on an open file, and another bonus to occupy the 7th rank is a typical example to mimic some kind of strategic planning. However, long-range planning and seeking for goals requires pattern recognition and associated advices if they match.

Oracle Approaches

Oracle approaches such as pre-processing and initializing piece-square tables based on features and pattern of the root may realize plans like minority attack and along with pawn structure and king placement, plans to attack the king. However, with todays typical search depths one has to consider path-dependencies, transposition table anomalies and resulting search instabilities. Instead, oracle approaches at or near the root, may be used to initialize dedicated piece-square tables for the sole purpose to control selectivity further down the tree, for instance to extend moves according to a certain plan.

Chess Quotes

[5]

  • Frank Marshall: A bad plan is better than none at all.
  • Emanuel Lasker: To find the right plan is just as hard as looking for its sound justification.
  • Eugene Znosko-Borovsky: It is not a move, even the best move that you must seek, but a realizable plan.
  • Alexander Kotov: It is better to follow out a plan consistently even if it isn't the best one than to play without a plan at all. The worst thing is to wander about aimlessly.

Chess Programs

See also

Publications

1946 ...

1950 ...

  • Max Euwe (1953). Judgment and Planning in Chess. McKay Company, New York, N.Y. [6]

1960 ...

1970 ...

1980 ...

1985 ...

1990 ...

1995 ...

2000 ...

2005 ...

2010 ...

2015 ...

2020 ...

Forum Posts

External Links

References

  1. Лозунг «План — закон, выполнение — долг, перевыполнение — честь!» в здании Ольховско-Батьковского торфяного предприятия в Кубринске, Переславский район - «Plan is law, fulfillment is duty, over-fulfillment is honor!» Building Olkhovskaya Batkovsky - peat enterprise Kubrinsk, Pereslavskiy district, Photo by Н. Частов (N. Part), November 1997, Five-Year Plans for the National Economy of the Soviet Union - Wikipedia
  2. Jan van Reek (1997). Strategy in Chess. Schachfirma Fruth, Unterhaching, ISBN 3-9804896-9-8
  3. Jan van Reek, Jos Uiterwijk, Jaap van den Herik (1998). Planning a Strategy in Chess. ICCA Journal, Vol. 21, No. 3
  4. Adriaan de Groot (1946). Het denken van den Schaker, een experimenteel-psychologische studie. Ph.D. thesis, University of Amsterdam; N.V. Noord-Hollandse Uitgevers Maatschappij, Amsterdam. Translated with the help of George Baylor, with additions, (in 1965) as Thought and Choice in Chess. Mouton Publishers, The Hague. ISBN 90-279-7914-6
  5. Chess Quotes - Strategy
  6. Judgment and Planning in Chess, games from Dr. Max Euwe's book Judgment and Planning in Chess. chessgames.com
  7. Re: Interesting ideas by Karlo Bala Jr., CCC, September 09, 2015
  8. New DeepMind paper by GregNeto, CCC, November 21, 2019
  9. MuZero: Mastering Go, chess, shogi and Atari without rules

Up one Level