Difference between revisions of "Dynamic Clustering"
Jump to navigation
Jump to search
m (fix mistakenly pural-ed word) |
m (a few typos / small fixes) |
||
| Line 1: | Line 1: | ||
| − | '''Dynamic clustering''' is a technique to find entries in your [[:Category:Log-Based Algorithms|log]] similar to the current situation. Essentialy, it is a [[wikipedia:K-nearest neighbor algorithm|K-nearest neighbor algorithm]], and not actually [[wikipedia:Cluster analysis|clustering]] at all. Despite this | + | '''Dynamic clustering''' is a technique to find entries in your [[:Category:Log-Based Algorithms|log]] similar to the current situation. Essentialy, it is a [[wikipedia:K-nearest neighbor algorithm|K-nearest neighbor algorithm]], and not actually [[wikipedia:Cluster analysis|clustering]] at all. Despite this misnomer, the term "Dynamic Clustering" has stuck with the robocode community. |
| − | The ideas is to record a "state" (or termed "situation") for each entry in your log. The state can contain any data that you | + | The ideas is to record a "state" (or termed "situation") for each entry in your log. The state can contain any data that you deem valuable, such as [[lateral velocity]], [[advancing velocity]], or [[enemy distance]]. Save this along with your data. Then to use the data, you find a "distance" between current state and past states. Distance can be [[wikipedia:Euclidean distance|Euclidian]] (<code>sqrt(sqr(dist1 - dist2) + sqr(lat1 - lat2))</code>) or another way, such as [[wikipedia:Manhattan distance|Manhattan distance]]. Find some number of entries with the lowest distance, and use them for targeting, movement, or whatever you like. |
| − | The earliest method doing this | + | The earliest method doing this was by iterating through the log and calculating the distance for each log entry. If you have a large log this is very slow. More recently [[Kd-Tree|Kd-Trees]] have been used. [[User:Corbos|Corbos]] was the first one to mention them on the [[Robowiki]], which caught the interest of [[User:Chase-san|Chase-san]] and [[User:Simonton|Simonton]]. As of now, Simonton's Kd-tree implementation is one of the faster ones. |
=== Bot using this technique === | === Bot using this technique === | ||
| − | * [[Tron]] & [[Shadow]]: First | + | * [[Tron]] & [[Shadow]]: First two bots using this technique. This technique used to called "Trons Gun", but later it was renamed. |
| − | * [[Chalk]]: First dynamic clustering bot that released with [[Open Source|source code]]. | + | * [[Chalk]]: First dynamic clustering bot that was released with [[Open Source|source code]]. |
| − | * [[DCBot]]: A simplified version of gun | + | * [[DCBot]]: A simplified version of gun used by earlier version of [[Tron]] and [[Shadow]]. |
* [[Lukious]] & [[Firebird]] & [[Hydra]]: Dynamic Clustering version of [[Dookious]], [[Phoenix]], [[WaveSerpent]] respectively. | * [[Lukious]] & [[Firebird]] & [[Hydra]]: Dynamic Clustering version of [[Dookious]], [[Phoenix]], [[WaveSerpent]] respectively. | ||
* [[Horizon]] & [[RougeDC]] & [[YersiniaPestis]]: Newer dynamic clustering bots. | * [[Horizon]] & [[RougeDC]] & [[YersiniaPestis]]: Newer dynamic clustering bots. | ||
Revision as of 04:34, 27 April 2009
Dynamic clustering is a technique to find entries in your log similar to the current situation. Essentialy, it is a K-nearest neighbor algorithm, and not actually clustering at all. Despite this misnomer, the term "Dynamic Clustering" has stuck with the robocode community.
The ideas is to record a "state" (or termed "situation") for each entry in your log. The state can contain any data that you deem valuable, such as lateral velocity, advancing velocity, or enemy distance. Save this along with your data. Then to use the data, you find a "distance" between current state and past states. Distance can be Euclidian (sqrt(sqr(dist1 - dist2) + sqr(lat1 - lat2))) or another way, such as Manhattan distance. Find some number of entries with the lowest distance, and use them for targeting, movement, or whatever you like.
The earliest method doing this was by iterating through the log and calculating the distance for each log entry. If you have a large log this is very slow. More recently Kd-Trees have been used. Corbos was the first one to mention them on the Robowiki, which caught the interest of Chase-san and Simonton. As of now, Simonton's Kd-tree implementation is one of the faster ones.
Bot using this technique
- Tron & Shadow: First two bots using this technique. This technique used to called "Trons Gun", but later it was renamed.
- Chalk: First dynamic clustering bot that was released with source code.
- DCBot: A simplified version of gun used by earlier version of Tron and Shadow.
- Lukious & Firebird & Hydra: Dynamic Clustering version of Dookious, Phoenix, WaveSerpent respectively.
- Horizon & RougeDC & YersiniaPestis: Newer dynamic clustering bots.
- X2 & Ali & DrussGT: These bots use Dynamic Clustering only for their gun.
