Smart bots competition

From RoboWiki
Fragment of a discussion from User talk:Beaming
Jump to: navigation, search

I like the idea of more processing time in order to attempt more complex stats, but to be honest I've tried some pretty complex stuff (including Spectral Clustering) for KNN on recorded data and nothing has beat simple KNN with a square kernel and weighted on sample distance. Not to mention that a lot of these algorithms aren't just a constant 50x slower or whatever, they essentially become unsolvable at sizes above ~500 data points, scaling quadratically or worse.

A problem with unlimited processing time is that it becomes possible to simply keep a copy of other bots and simulate them to determine where they will shoot/move.

Does anybody have any specific techniques that they would use if there was more processing time available?

Skilgannon (talk)17:03, 13 October 2014

I like the idea of more processing time in order to attempt more complex stats, but to be honest I've tried some pretty complex stuff (including Spectral Clustering) for KNN on recorded data and nothing has beat simple KNN with a square kernel and weighted on sample distance. Not to mention that a lot of these algorithms aren't just a constant 50x slower or whatever, they essentially become unsolvable at sizes above ~500 data points, scaling quadratically or worse.

A problem with unlimited processing time is that it becomes possible to simply keep a copy of other bots and simulate them to determine where they will shoot/move.

Does anybody have any specific techniques that they would use if there was more processing time available?

Beaming (talk)17:23, 13 October 2014

I like the idea of more processing time in order to attempt more complex stats, but to be honest I've tried some pretty complex stuff (including Spectral Clustering) for KNN on recorded data and nothing has beat simple KNN with a square kernel and weighted on sample distance. Not to mention that a lot of these algorithms aren't just a constant 50x slower or whatever, they essentially become unsolvable at sizes above ~500 data points, scaling quadratically or worse.

A problem with unlimited processing time is that it becomes possible to simply keep a copy of other bots and simulate them to determine where they will shoot/move.

Does anybody have any specific techniques that they would use if there was more processing time available?

Beaming (talk)23:23, 13 October 2014
 

I like the idea of more processing time in order to attempt more complex stats, but to be honest I've tried some pretty complex stuff (including Spectral Clustering) for KNN on recorded data and nothing has beat simple KNN with a square kernel and weighted on sample distance. Not to mention that a lot of these algorithms aren't just a constant 50x slower or whatever, they essentially become unsolvable at sizes above ~500 data points, scaling quadratically or worse.

A problem with unlimited processing time is that it becomes possible to simply keep a copy of other bots and simulate them to determine where they will shoot/move.

Does anybody have any specific techniques that they would use if there was more processing time available?

Rednaxela (talk)17:36, 13 October 2014
 

I like the idea of more processing time in order to attempt more complex stats, but to be honest I've tried some pretty complex stuff (including Spectral Clustering) for KNN on recorded data and nothing has beat simple KNN with a square kernel and weighted on sample distance. Not to mention that a lot of these algorithms aren't just a constant 50x slower or whatever, they essentially become unsolvable at sizes above ~500 data points, scaling quadratically or worse.

A problem with unlimited processing time is that it becomes possible to simply keep a copy of other bots and simulate them to determine where they will shoot/move.

Does anybody have any specific techniques that they would use if there was more processing time available?

Chase17:58, 13 October 2014
 
Personal tools