From Robowiki
Jump to navigation Jump to search

From old wiki's BestPSpace page

My prediction: This is the firing that the future killer bot will use. --tobe

Seems like it might have problems with learning speed, especially if the eternal rumble continues to use 10 round battles. --David Alves

So was this basically segmentation of GF targeting, or is there more to it? -- Jokester

When I read it I understood it as GF segmentation, but with segments treated separately. Maybe I was wrong ;) --lRem

Seems like a good system. The real trick to making it learn faster is to choose families that show some real relationships. You could have many different options for families built into the code, but only store and collect the ones that show a significant correlation (thankfully, there are established ways to evaluate this). I would be willing to bet that the biggest source of problems with this method is the fact that 90% of top bots use some form of wave surfing, so only the GuessFactorTargeting family will show any correlation. --speal


Thread titleRepliesLast modified
Relation to "many vcs buffers"105:40, 10 July 2020

Relation to "many vcs buffers"

I think this method has some relation with the popular many VCS buffers way, in which each VCS buffer forms a probability space, and by accumulating, the best group of probability space is chosen, and when having less data, it downgrades automatically.

k-nn with many dimensions does similar thing, only smoother imo.

Xor (talk)05:38, 10 July 2020

However, in wave surfing, it's still in question whether something in the k-nn world can match the performance of a hundred VCS buffers like in DrussGT ;)

maybe some combination of BestPSpace and k-nn will work in wave surfing?

Xor (talk)05:40, 10 July 2020