Difference between revisions of "Summary:Talk:Dynamic Clustering/Using previous GFs as dimensions"

From Robowiki
Jump to navigation Jump to search
(Summary via ChatGPT)
 
m (A better summary via ChatGPT)
 
Line 1: Line 1:
−
A group of robocode enthusiasts are discussing a gun that uses kNN classification, with one of the dimensions being the last five visited GuessFactors (GFs) by the opponent. They discuss testing the gun against different movement styles of robots, including Random Movement (RM), and using various types of predictors. They also discuss the number of dimensions, data decay, and the order in which the last five GFs are fed to the gun. They consider the possibility of using a different clustering method instead of kNN, and test the gun against DrussGT's flattener, achieving comparable results to Diamond's gun. They also discuss the impact of increasing K in GFs as dimensions.
+
Straw is discussing his development of a gun for a Robocode competition that uses kNN classification and the opponent's last 5 GFs (guess factor) as dimensions. The gun performs well against surfers and simple movers, but has not been tested against random movement. Rednaxela suggests using the reference bots from the Targeting Challenge RM for testing. The gun is still being optimized and Straw is considering using a VG array or density calculations. Voidious is curious if the 5 GFs are treated independently or as a sequence. Rednaxela suggests using relative GFs as dimensions. Straw also mentions using 10 GFs instead of 5 and using descending weights on older GFs. The gun has shown promising results against flatteners, with a similar hit rate as Diamond's gun. The performance of the gun improves with an increased value of K.
  
 
– ChatGPT
 
– ChatGPT

Latest revision as of 17:27, 12 February 2023

Straw is discussing his development of a gun for a Robocode competition that uses kNN classification and the opponent's last 5 GFs (guess factor) as dimensions. The gun performs well against surfers and simple movers, but has not been tested against random movement. Rednaxela suggests using the reference bots from the Targeting Challenge RM for testing. The gun is still being optimized and Straw is considering using a VG array or density calculations. Voidious is curious if the 5 GFs are treated independently or as a sequence. Rednaxela suggests using relative GFs as dimensions. Straw also mentions using 10 GFs instead of 5 and using descending weights on older GFs. The gun has shown promising results against flatteners, with a similar hit rate as Diamond's gun. The performance of the gun improves with an increased value of K.

– ChatGPT

Retrieved from "http://robowiki.net/w/index.php?title=Summary:Talk:Dynamic_Clustering/Using_previous_GFs_as_dimensions&oldid=57346"