Content deleted Content added
m Dating maintenance tags: {{Dead link}} |
m linking |
||
Line 6:
One method of choice is embodied in [[backtracking]] systems (such as [[Amb (evaluator)|Amb]],<ref>{{Cite web|url=https://mitpress.mit.edu/sites/default/files/sicp/full-text/book/book-Z-H-28.html#%_sec_4.3.3|title=Structure and Interpretation of Computer Programs}}{{dead link|date=April 2023}}</ref> or unification in [[Prolog]]), in which some alternatives may "fail," causing the program to backtrack and try other alternatives. If all alternatives fail at a particular choice point, then an entire branch fails, and the program will backtrack further, to an older choice point. One complication is that, because any choice is tentative and may be remade, the system must be able to restore old program states by undoing side-effects caused by partially executing a branch that eventually failed.
Another method of choice is [[reinforcement learning]], embodied in systems such as [[Alisp]].<ref>{{cite journal|author1=David Andre |author2=Stuart J. Russell|title=State abstraction for programmable reinforcement learning agents|journal=Eighteenth National Conference on Artificial Intelligence|date=July 2002|pages=119–125|url=https://dl.acm.org/doi/10.5555/777092.777114}}</ref> In such systems, rather than backtracking, the system keeps track of some measure of success and learns which choices often lead to success, and in which situations (both internal program state and environmental input may affect the choice). These systems are suitable for applications to [[robotics]] and other domains in which backtracking would involve attempting to undo actions performed in a dynamic environment, which may be difficult or impractical.
==See also==
| |||