Difference between pages "GitWikiBot TestPage" and "Raven"

From Robowiki
(Difference between pages)
Jump to navigation Jump to search
(testing upload from file summary)
 
(Raven page update)
 
Line 1: Line 1:
−
{{Navbox small
+
== Background Information ==
−
| title = Wraith Sub-pages
+
 
−
| parent = Wraith
+
; Bot Name
−
| page1 = Version History
+
: Raven
−
| page2 = Challenge Results
 
−
| page3 = Wolfman's TODO List
 
−
}}
 
  
−
Wraith is a refactor of my bot [[AgentSmithRedux]].
+
; Author
 +
: [[Dsekercioglu]]
  
−
{{Infobox Robot
+
; Extends
−
| author          = [[User:Wolfman|Wolfman]]
+
: [[AdvancedRobot]]
−
| extends        = [[AdvancedRobot]]
 
−
| targeting      = Virtual Guns with Linear, Circular and Head On Targeting
 
−
| movement        = [[DangerPrediction]]
 
−
| current_version = 0.1
 
−
| download_link  = https://drive.google.com/uc?export=download&id=1MlGa-docshYLFIc2udl5qOyXdT4xL6KF
 
−
| isOpenSource    = no
 
−
| isOneOnOne      = yes
 
−
| isMelee        = not yet but planned
 
−
}}
 
  
−
== Background Information ==
+
; What's special about it?
 +
: Raven dynamically weights its movement classifiers using gradient-based optimization.
 +
: Raven uses a form of Go to surfing where it procedurally generates paths without aiming for any point.
  
−
Returning after 3 years, I wanted to have a bit of fun and restart some robocode projects. Taking [[AgentSmithRedux]] and improving upon it for fun.
+
; Great, I want to try it. Where can I download it?
 +
: https://www.dropbox.com/s/335p9lcdwgpagry/dsekercioglu.mega.Raven_3.19b.jar?dl=1
  
−
It is currently in very early testing and is not expected to be competitive till it has all its intended features & iteration.
+
; How competitive is it?
 +
: Its best is 15.
  
 +
; Credits
 +
: [[Rednaxela]], [[Skilgannon]], [[Nat]], [[Starrynyte]] and the other contributors I am unaware of for the FastTrig class.
 +
: [[Skilgannon]] for the bugless, fast Kd-tree.
 +
: [[Cb]] for the non-iterative wall smoothing.
 +
: [[Rozu]] for the precise prediction code.
 +
: [[Chase-san]] for the intercept method I used in my [[PPMEA]] calculations.
 +
: [[AW]] for giving me the idea of integrating the danger function in order to get the danger value for a given bot width.
 
== Strategy ==
 
== Strategy ==
  
−
;How does it [[:Category:Movement|move]]?
+
; How does it [[Movement|move]]?
 +
: A form of Go To Surfing.
 +
: It falls back to True Surfing when there is no bullets mid-air.
  
−
An evolution of the technique I've currently called [[DangerPrediction]] from [[AgentSmith]] which is a form of minimum risk movement whereby it plans its route ahead based on incoming predicted bullet positions from the enemies. It should theoretically work exactly the same in 1v1 and melee as the only difference is more incoming bullets in melee to avoid. Still plan to do a write up of [[DangerPrediction]] once its been nailed down and is competitive in 1v1.
+
; How does it fire?
 +
: It uses [[GuessFactor]] with KNN.
  
−
At the time of writing it will dodge HOT & linear guns at around 99+% efficiency, but does not yet take into account more advanced techniques that could be fired at it.
+
; How does it [[Dodging Bullets|dodge bullets]]?
 +
: Tries to minimize the number of guess factors it gets hit by based on their weights and damage.
  
−
;How does it [[:Category:Targeting|fire]]?
+
; What does it save between rounds and matches?
−
 
+
: Between rounds, it saves the kd-trees. Between matches, it doesn't save anything.
−
Its currently just got a simple Virtual Gun array at the moment choosing the best gun between various guns including Linear, Circular and Head On.
 
−
 
 
−
;How does it [[Dodging Bullets|dodge bullets]]?
 
−
 
 
−
See Movement above!
 
  
 
== Additional Information ==
 
== Additional Information ==
  
−
;Where did you get the name?
+
; Where did you get the name?
−
 
+
: It just popped into my mind and I thought it would be a proper name for a bot with machine learning.
−
Wraiths can only be hit by magical weapons! ;)
 
  
−
== Credits ==
+
; Can I use your code?
 +
: Yes, I tried to make the code as clean and understandable as possible.
  
−
Currently contains the source code for Raiko Micro's gun, but it is disabled as it was required for MC2K7 challenge testing.
+
; What's next for your robot?
 +
: A proper versioning system so I don't keep accidentally releasing experimental versions.
 +
: Faster code so it doesn't run slow and doesn't skip turns.
 +
: Better bullet shadow calculations.
 +
: Tuning the guns since they haven't been tuned since the first version.
 +
: Gun Heat Waves.
 +
: Maybe a pre-trained movement or gun to use in the first ticks of the battle.
 +
: Add a flattener that actually improves its scores against adaptive targeting.
  
−
== Thanks To ==
+
; Does it have any [[White Whale]]s?
 +
: Raven doesn't seem to have any extreme white whales currently.
  
−
Everyone who contributes to the RoboWiki. It's an awesome resource, be proud of yourselves!
+
; What other robot(s) is it based on?
 +
: It's kind of based on WhiteFang, I have tried to copy the design but make it as precise as it can be.
  
−
[[Category:1-vs-1_Bots|Wraith]]
+
[[Category:Templates|Bots]]

Revision as of 15:26, 27 July 2020

Background Information

Bot Name
Raven
Author
Dsekercioglu
Extends
AdvancedRobot
What's special about it?
Raven dynamically weights its movement classifiers using gradient-based optimization.
Raven uses a form of Go to surfing where it procedurally generates paths without aiming for any point.
Great, I want to try it. Where can I download it?
https://www.dropbox.com/s/335p9lcdwgpagry/dsekercioglu.mega.Raven_3.19b.jar?dl=1
How competitive is it?
Its best is 15.
Credits
Rednaxela, Skilgannon, Nat, Starrynyte and the other contributors I am unaware of for the FastTrig class.
Skilgannon for the bugless, fast Kd-tree.
Cb for the non-iterative wall smoothing.
Rozu for the precise prediction code.
Chase-san for the intercept method I used in my PPMEA calculations.
AW for giving me the idea of integrating the danger function in order to get the danger value for a given bot width.

Strategy

How does it move?
A form of Go To Surfing.
It falls back to True Surfing when there is no bullets mid-air.
How does it fire?
It uses GuessFactor with KNN.
How does it dodge bullets?
Tries to minimize the number of guess factors it gets hit by based on their weights and damage.
What does it save between rounds and matches?
Between rounds, it saves the kd-trees. Between matches, it doesn't save anything.

Additional Information

Where did you get the name?
It just popped into my mind and I thought it would be a proper name for a bot with machine learning.
Can I use your code?
Yes, I tried to make the code as clean and understandable as possible.
What's next for your robot?
A proper versioning system so I don't keep accidentally releasing experimental versions.
Faster code so it doesn't run slow and doesn't skip turns.
Better bullet shadow calculations.
Tuning the guns since they haven't been tuned since the first version.
Gun Heat Waves.
Maybe a pre-trained movement or gun to use in the first ticks of the battle.
Add a flattener that actually improves its scores against adaptive targeting.
Does it have any White Whales?
Raven doesn't seem to have any extreme white whales currently.
What other robot(s) is it based on?
It's kind of based on WhiteFang, I have tried to copy the design but make it as precise as it can be.
Retrieved from "http://robowiki.net/w/index.php?title=GitWikiBot_TestPage&oldid=56295"