GsocAIProjectProgress - Revision history

Dhains: /* WML Tags */

2008-05-26T17:59:30Z

‎WML Tags

Dhains: /* WML Tags */

2008-05-26T17:59:17Z

‎WML Tags

Dhains: /* WML Tags */

2008-05-26T17:49:18Z

‎WML Tags

Aethaeryn: categorizing

2008-05-17T13:13:09Z

categorizing

Dhains: /* AI API: Formula AI -> WML -> AI */

2008-04-30T19:01:48Z

‎AI API: Formula AI -> WML -> AI

Dhains: /* WML Tags */

2008-04-30T19:01:08Z

‎WML Tags

Dhains: /* WML Tags */

2008-04-30T18:48:24Z

‎WML Tags

Dhains: /* AI API: Formula AI -> WML -> AI */

2008-04-30T18:48:06Z

‎AI API: Formula AI -> WML -> AI

Dhains: /* Basic Idea */

2008-04-29T15:19:53Z

‎Basic Idea

Dhains: /* Basic Idea */

2008-04-29T15:17:57Z

‎Basic Idea

@@ Line 58: / Line 58: @@
    name= <name of move>
    type= movement/recruitment/etc
-   result= <action returning formula>
+   action= <action returning formula>
    evaluation= <int returning formula>
   [/register_candidate_move]

@@ Line 58: / Line 58: @@
    name= <name of move>
    type= movement/recruitment/etc
    evaluation= <int returning formula>
   [/register_candidate_move]

@@ Line 56: / Line 56: @@
   [register_candidate_move]
    type= movement/recruitment/etc
    evaluation= <int returning formula>

← Older revision		Revision as of 13:13, 17 May 2008
Line 64:		Line 64:

	Here we define how the Formula AI interfaces with the C++ part (i.e. how should moves be supplied, how the moves will be selected and weights applied)		Here we define how the Formula AI interfaces with the C++ part (i.e. how should moves be supplied, how the moves will be selected and weights applied)
		+
		+	[[Category:Summer of Code]]

@@ Line 45: / Line 45: @@
 Here we define how Formula AI should be written (Formula AI scripts, Formula AI libraries, etc.)
-=== AI API: Formula AI -> WML -> AI ===
+=== AI API: Formula AI & WML  ===
 Here we define how the formula AI will be included in WML and how the WML will be processed.  This layer should be transparent to the AI itself, but will be the one mainly used by scenario designers and AI designers.

@@ Line 51: / Line 51: @@
 ==== WML Tags ====
-Currently formula can be included with formula=.  The AI will need to know what the prupose of some formula is, i.e. we might have special formula for recruitment, evaluation functions, candidate move selection, etc.  Additional formula inclusion tags will be used to classify these different formula function types
+Currently formula can be included with formula=.  The AI will need to know what the purpose of some formula is, i.e. we might have special formula for recruitment, candidate moves selection, etc.  We also wish to return an integer which would be an evaluation for the proposed move.
-* formula_recruit : this tag will specify a formula to be used for recruitment
+ [register_candidate_move]
-* formula_eval : This tag specifies evaluation functions
+  type= movement/recruitment/etc
-* move: formula_move : Candidate moves (?), we might not need this, it may be OK to just use the normal formula tag
+  evaluation= <int returning formula>
 === AI API: Formula AI -> C++ ===
 Here we define how the Formula AI interfaces with the C++ part (i.e. how should moves be supplied, how the moves will be selected and weights applied)

@@ Line 9: / Line 9: @@
 Eval functions in FormulaAI will be provided which will associate a 'score' to each unit the AI can recruit.  Based on these scores, the AI will recruit particular units.  Along with each of these 'scores' is an associated weight, originally set to 1.  Now suppose due fog of war or some other event, the AI had no knowledge of an enemy tactic or unit and a certain unit which the Formula AI ranked very high begins to be massacred due to this incomplete knowledge.  The weight associated with that particular unit can now be adjusted to some value < 1 by the AI.  This will decrease the chance of the AI choosing these units, even though the FormulaAI scores them fairly high.
-The point of this learning system is that deterministic rules maybe good for many situations, but they cannot predict everything.  In fact, experienced players may infer some of the deterministic rules and use them to their advantage.  The learning system will give the AI the ability to tweak it's strategy on the fly to cope with player strategies that the deterministic rules cannot deal with.  Recruitment is aless complex then a full fledged AI and the learning system may not be as necessary as it will be in a full fledged AI, but recruitment serves as a useful illustration.
+The point of this learning system is that deterministic rules are sufficient for many situations, but they cannot predict everything.  There will always be some area of incomplete knowledge, i.e. we cannot code into the deterministic rules complete knowledge of the game and every situation the AI will ever face.  In fact, experienced players may even be able to infer some of the deterministic rules and use this knowledge to their advantage.  The learning system will give the AI the ability to tweak it's strategy on the fly to cope with player strategies that the deterministic rules cannot deal with.  Recruitment is less complex then a full fledged AI and the learning system may not be as necessary as it will be in a full fledged AI, but recruitment serves as a useful illustration.
 === Questions ===

@@ Line 7: / Line 7: @@
 For example, consider a recruitment example:
-Eval functions in FormulaAI will be provided which will associate a 'score' to each unit the AI can recruit.  Based on these scores, the AI will recruit particular units.  Along with each of these 'scores' is an associated weight, originally set to 1.  Now suppose due fog of war or some other event the AI had no knowledge of certain players tactics or units and a certain unit which the Formula AI ranked very high begins to be massacred.  Now the weight associated with that particular unit can be adjusted to some value < 1.  This will decrease the chance of the AI choosing these units, even though the FormulaAI scores them fairly high.
+Eval functions in FormulaAI will be provided which will associate a 'score' to each unit the AI can recruit.  Based on these scores, the AI will recruit particular units.  Along with each of these 'scores' is an associated weight, originally set to 1.  Now suppose due fog of war or some other event, the AI had no knowledge of an enemy tactic or unit and a certain unit which the Formula AI ranked very high begins to be massacred due to this incomplete knowledge.  The weight associated with that particular unit can now be adjusted to some value < 1 by the AI.  This will decrease the chance of the AI choosing these units, even though the FormulaAI scores them fairly high.
 The point of this learning system is that deterministic rules maybe good for many situations, but they cannot predict everything.  In fact, experienced players may infer some of the deterministic rules and use them to their advantage.  The learning system will give the AI the ability to tweak it's strategy on the fly to cope with player strategies that the deterministic rules cannot deal with.  Recruitment is aless complex then a full fledged AI and the learning system may not be as necessary as it will be in a full fledged AI, but recruitment serves as a useful illustration.