|
Discussion moved from more specific page ... |
|
One of the problems with StatisticalTargeting is dealing with multi-mode bots like PrairieWolf and Ares. Once you have gotten some data in your stat buffers on these bots they tend to switch mode and your data is useless. After a while you have some average data that probably isn't very effective against any of the modes of the enemy. One way of dealing with this problem is by adding some decay to your stats so that you "forget" older data. The most well known approach to this is RollingAverages introduced by Paul Evans with his SandboxLump eons ago in Robocode time. If you are using VisitCountStats you might get some decay effects out of using VisitCountStats/LimitFileSize. Other schemes might involve regularly dividing all stat buckets with some value to keep learnt lessons easies to unlearn. |
|
Much of the following discussion should probably move to other (probably new) pages... |
|
I think something like DT's adaptive movement, PrairieWolf's multi-mode movement, and a couple unchanging movements (maybe one bullet-dodging and one not) would be the way to go about testing it. -- Kawigi |
|
I think something like DT's adaptive movement, PrairieWolf's multi-mode movement, and a couple unchanging movements (maybe one bullet-dodging and one not) would be the way to go about testing it. -- Kawigi |
![[Home]](/images/RoboWiki.png)