Goals and planning - criticisms


next up previous contents
Next: Criticisms
Up: No Title
Previous: Introduction - criticisms of AI
Back: to main list of student notes

Goals and planning - criticisms

In this section, I'll focus on the notion of goal, and I'll trace a trajectory through ``criticism space'', ending up with the complete abolition of goals and even computation.

In the early 60s, Newell Simon and Shaw introduced the General Problem Solver, GPS. This marked the start of the planning paradigm. I discuss GPS in my second set of tutorial notes. For now, all you need to know is that GPS sees problems in terms of goals to be achieved. Each goal describes what state the world must be in for the problem to have been accomplished. GPS also has a table of operators. These are the basic actions available to the problem solver: the table specified how each action, if carried out alone, transforms the state of the world. Using a means-ends algorithm, GPS then attempts to find a sequence of actions which transforms the initial state of the world into one of the goal states.

For references to GPS, Boden gives a clear non-technical account in pages 354-357 of AI and Natural Man. Winston has a more detailed account, with examples of how one formalises problems. One of the original articles appears in GPS, A Program that Simulates Human Thought, by A Newell and H A Simon, in Computers and Thought edited by E A Feigenbaum and J Feldman (RSL: per 19668 d 335; PSY: KH:F 032). In the latter article, the authors state that they believe the subjects they tested were manipulating symbols in the same way as GPS (and hence presumably had an explicit representation for their goals). This article also appears in Readings in Planning (page 59), location below.




next up previous contents
Next: Criticisms
Up: No Title
Previous: Introduction - criticisms of AI
Back: to main list of student notes



Jocelyn Ireson-Paine
Wed Feb 14 23:51:11 GMT 1996