User:Dragos/GsoC 2013/AI: Implement a 'total defense' strategy - Revision history

Dragos at 14:11, 3 May 2013

2013-05-03T14:11:38Z

Dragos at 22:33, 30 April 2013

2013-04-30T22:33:36Z

Dragos at 21:35, 29 April 2013

2013-04-29T21:35:27Z

Dragos at 01:43, 29 April 2013

2013-04-29T01:43:23Z

Dragos at 10:09, 28 April 2013

2013-04-28T10:09:05Z

Dragos at 01:03, 25 April 2013

2013-04-25T01:03:02Z

Dragos at 22:35, 22 April 2013

2013-04-22T22:35:39Z

Dragos at 22:27, 22 April 2013

2013-04-22T22:27:28Z

Dragos at 22:21, 22 April 2013

2013-04-22T22:21:24Z

Show changes

Crab at 12:39, 22 April 2013

2013-04-22T12:39:15Z

@@ Line 1: / Line 1: @@
 {{SoC2013Student}}
 [[Category:SoC Ideas AI Defense Strategies 2013]]
 =Description=
@@ Line 14: / Line 11: @@
 =Project=
 ==Make the AI do a ‘full retreat’==
@@ Line 37: / Line 36: @@
 The changes will need extensive tests. We will need tests against AI and against real players so we can be sure the AI improves overall effectiveness and to be 100% sure we leave no loopholes that will make the AI easy to trick/defeat.
 =Questionnaire=
@@ Line 175: / Line 173: @@
 This project combines thinking of a solution (and I consider that this will probably be the hardest part in this project) with working at a good implementation. I think that this mix of practical and theoretical work is perfect for my skills.
-===Include an estimated timeline for your work on the project. Don't forget to mention special things like "I booked holidays between A and B" and "I got an exam at ABC and won't be doing much then".===
+===Include an estimated timeline for your work on the project. Include as much technical detail about your implementation as you can===
 {| class="wikitable"
@@ Line 213: / Line 211: @@
 . There is a possibility to add a similar testing phase after the midterm evaluation, to see if we need to focus more time on some aspects of the project. I emphasized so much on testing because I believe that it’s hard to predict how the AI will act while developing it.
 ===What do you expect to gain from this project?===

@@ Line 181: / Line 181: @@
 ! width="300pt"| Time Frame !! Milestone
 |-
-| 14 June - 24 June (11 days) || Implementing the full retreat function and testing formulas for the positions
+| 14 June - 23 June (10 days) || Implementing the full retreat function and testing formulas for the positions
 |-
-| 4 June - 28 June (5 days) || Testing the full retreat and analyzing the available formulas to choose the best one
+| 24 June - 27 June (4 days) || Testing the full retreat and analyzing the available formulas to choose the best one
 |-
-| 28 June - 11 July (15 days) || Writing an algorithm to make the units group themselves when necessary
+| 28 June - 10 July (14 days) || Writing an algorithm to make the units group themselves when necessary
 |-
 | 11 July - 14 July (4 days) || Testing the grouping algorithm

← Older revision		Revision as of 21:35, 29 April 2013
Line 170:		Line 170:

	I’ve selected improving the AI, more specifically implementing a ‘total defense’ strategy.		I’ve selected improving the AI, more specifically implementing a ‘total defense’ strategy.
−
−	~~===If you have invented your own project, please describe the project and the scope.===~~

	===Why did you choose this project?===		===Why did you choose this project?===

← Older revision		Revision as of 01:43, 29 April 2013
Line 178:		Line 178:

	===Include an estimated timeline for your work on the project. Don't forget to mention special things like "I booked holidays between A and B" and "I got an exam at ABC and won't be doing much then".===		===Include an estimated timeline for your work on the project. Don't forget to mention special things like "I booked holidays between A and B" and "I got an exam at ABC and won't be doing much then".===
		+
		+	{\| class="wikitable"
		+	\|-
		+	! width="300pt"\| Time Frame !! Milestone
		+	\|-
		+	\| 14 June - 24 June (11 days) \|\| Implementing the full retreat function and testing formulas for the positions
		+	\|-
		+	\| 4 June - 28 June (5 days) \|\| Testing the full retreat and analyzing the available formulas to choose the best one
		+	\|-
		+	\| 28 June - 11 July (15 days) \|\| Writing an algorithm to make the units group themselves when necessary
		+	\|-
		+	\| 11 July - 14 July (4 days) \|\| Testing the grouping algorithm
		+	\|-
		+	\| 15 July - 4 August (21 days) \|\| Refactoring combat evaluation code and working to take into account the next turn. (In this period I will also prepare for the midterm evaluation)
		+	\|-
		+	\| 5 August- 8 August (4 days) \|\| Testing the next turn and the modifications to the combat evaluation
		+	\|-
		+	\| 9 August- 23 August (15 days) \|\| Adding better usage for support units
		+	\|-
		+	\| 24 August - 24 August (1 day) \|\| Testing the support units individually and all together
		+	\|-
		+	\| 25 August - 10 September (17 days) \|\| Implementing ideas that may come during the time I work on the project. If there aren’t any new ideas I will reallocate the time as necessary
		+	(See notes 1 and 3)
		+	\|-
		+	\| 11 September - 23 September (13 days) \|\| Extensive testing of all the ideas applied together, with help from Wesnoth players (if possible), preparing for the final evaluation, writing necessary documentation
		+	\|-
		+	\| After SOC (PLAN) \|\| Continue working for Wesnoth
		+	\|}
		+
		+	Notes:
		+
		+	1. I’ve allocated some time from 25 August to 10 September to implement new ideas if they come. In the case that I have no other ideas, I will reallocate the time for the ideas that need it more and/or I will have an after midterm testing phase, like the one at the end of the project (Note 3).
		+
		+	2. After each implementation, I will use a couple of days to test the implementation formula and to search for bugs. The last testing phase, that takes a week, will test all the combined ideas working together to see how the AI behaves. This testing phase will (hopefully) be done by me and other people from the Wesnoth community. I hope the last testing phase will help tweak the existing ideas, to make them perform even better.
		+
		+	3. There is a possibility to add a similar testing phase after the midterm evaluation, to see if we need to focus more time on some aspects of the project. I emphasized so much on testing because I believe that it’s hard to predict how the AI will act while developing it.

	===Include as much technical detail about your implementation as you can===		===Include as much technical detail about your implementation as you can===

@@ Line 8: / Line 8: @@
 <h4>Dragos Ristache: Implement a total defense strategy </h4>
-I intent to implement a total defense strategy for the Wesnoth AI. Currently the Wesnoth AI is really good at doing attacks but it's not that good at defending. The AI currently tries to maximize the formula (enemy losses - aggression*own_losses), which can't generate a defending behavior making it easy to defeat in certain situation.
+I intend to implement a total defense strategy for the Wesnoth AI. Currently the Wesnoth AI is really good at doing attacks but it's not that good at defending. The AI currently tries to maximize the formula (enemy losses - aggression*own_losses), which can't generate a defending behavior making it easy to defeat in certain situation.
 =IRC=

@@ Line 9: / Line 9: @@
 I intent to implement a total defense strategy for the Wesnoth AI. Currently the Wesnoth AI is really good at doing attacks but it's not that good at defending. The AI currently tries to maximize the formula (enemy losses - aggression*own_losses), which can't generate a defending behavior making it easy to defeat in certain situation.
 =IRC=
 dragos
 =Questionnaire=
@@ Line 158: / Line 182: @@
 ===What do you expect to gain from this project?===
 ===What would make you stay in the Wesnoth community after the conclusion of SOC?===
 ==Practical considerations==

@@ Line 4: / Line 4: @@
 =Description=
 <h4>Dragos Ristache: Implement a total defense strategy </h4>
-TODO: Write a small (1-4 sentences) description of your proposal here.
 TODO: Add more first-level sections to detail your proposal

← Older revision		Revision as of 12:39, 22 April 2013
Line 1:		Line 1:
	{{SoC2013Student}}		{{SoC2013Student}}
−	~~[[Category:SoC_Ideas_Your_Own_Ideas2013]]~~
	[[Category:SoC Ideas AI Defense Strategies 2013]]		[[Category:SoC Ideas AI Defense Strategies 2013]]