3.1.3DC vs 3.1.3

Jump to navigation Jump to search

Correct :-) In the movement, to be more specific.

And considering I write a changelog, I don't see how this question is anything other than lazy at its worst.

Skilgannon (talk)07:04, 25 December 2013

I was always wondering why the best bot used VCS, DC seems much more elegant. Does it improve performance in your tests?

Straw (talk)08:05, 25 December 2013

For some reason I've never managed to get the DC to perform as well as the VCS, so it still used VCS. I remember Jdev commenting that a range search worked better for him than a KNN search in movement, so I'll be trying that next.

Skilgannon (talk)21:34, 25 December 2013

Have you tried doing something similar to your many randomized attribute buffers with kD-Trees? You could make 100 trees, each with a random subset of the predictors, then combine the results. You could even start weighting some tree's results higher if they perform better.

Straw (talk)00:36, 14 January 2014

In my mind, a tree is more heavyweight than a buffer. You need multiple buffers to begin to approximate the smoothing you get from KNN and kernel density. I use multiple trees in my surfing for the same reasons, but more on the order of 10 than 100.

Voidious (talk)01:03, 14 January 2014
 

I was running some tests over the holidays, and it seems I've improved my DC to the point where against weaker bots it is as good as VCS. However, against top bots VCS movement is still much stronger, ~15% difference on the MC2K7.

Skilgannon (talk)07:08, 14 January 2014

How did you improve it?

Tmservo (talk)13:19, 14 January 2014

Adjusting rolling speed and changing the shape functions of some attributes.

Skilgannon (talk)20:17, 14 January 2014

What is rolling speed and shape functions

Tmservo (talk)23:51, 14 January 2014

Rolling speed is part of his moving average algorithm, basically how much new values supersede older ones based on time.

I assume shape functions here he means the shape of the kernel function he uses.

Chase08:18, 15 January 2014

Close. The shape functions are the nonlinear scalings I do on attributes before adding them to the tree.

Skilgannon (talk)08:43, 15 January 2014

Oooh, that makes more sense. :)

Chase17:23, 15 January 2014
 

Shape functions are part of the distance function then.

MN (talk)13:09, 16 January 2014
 
 
 
 
 
Edited by author.
Last edit: 08:41, 9 February 2014

You do not have permission to edit this page, for the following reasons:

  • The action you have requested is limited to users in the group: Users.
  • You must confirm your email address before editing pages. Please set and validate your email address through your user preferences.

You can view and copy the source of this page.

Return to Thread:Talk:DrussGT/Version History/3.1.3DC vs 3.1.3/reply (20).

Please go away.

Chase15:55, 17 January 2014
 
 

Could someone explain why averaging the results from many random trees is stronger than using a single well-tuned tree?

MN (talk)13:13, 16 January 2014

I would suspect it might make your nearest-neighbours come from multiple perspectives, giving you areas of concavity in your nearest-neighbour function instead of just a pure convex search area. I also suspect using some fancy pre-processing on tree attributes (perhaps dimension reduction/PCA) before adding could give equivalent search patterns.

Skilgannon (talk)13:56, 16 January 2014
 

I'd answer this in 3 parts.

  1. There are some high level movement classes that are worth segmenting. Against simple targeters, time since velocity change is just noise. Against most bots, a flattener would be noise. But for a bot where a flattener helps, those lower levels of stats don't hurt. I think they even add "harmless noise" - they are still bullet dodging, so they won't make horrible decisions. So I have a few tiers (simple, normal / decaying, light flattener, flattener) in my movement stats, enabled at different enemy hit percentages.
  2. I found VCS to be easier to tune than DC. Similarly, I think layering a few trees is easier than trying to add features to your KNN system to create the exact "shapes" (or however you imagine it) that you want. "5 of last 150 + 5 of last 500 + 5 of last 1500" is easy to understand. Adjusting the weights and distancing to produce the same results from one KNN call seems hard.
  3. I can't prove that it is.
Voidious (talk)16:50, 16 January 2014
 

I believe that the trees tend to cancel each others errors and over fitting. That is why Random Forest works. In addition,like Voidious said, simpler trees may be better vs simple enemies.

Straw (talk)19:37, 16 January 2014
 

But then each tree should be specifically tuned against a specific kind of gun. Then each tree outputs a spike at a different GF, which shouldn't be a problem since you can dodge many GFs at once.

But generating dimensions at random to mimic DrussGT 100 buffers is another matter entirely. A combination of dimensions which don't relate to any gun is supposed to hurt classification. Although I can't prove it either.

MN (talk)01:33, 17 January 2014

I'd tend to expect that when the "correct" parameters of the model (i.e. weightings of dimensions) are have more uncertainty than is in the resulting prediction of any one model, the consensus among a diverse set of models is less likely to be completely wrong than any one model. Or to put it another way, perhaps there there is no single well-tuned tree that fits all opponents of a large-ish category (i.e. "specific kind of gun") well enough to outperform a consensus of different models, and while there may exist well-tuned trees for smaller categories of opponents, the battles might not be long enough to reliably detect which would be the best category. That's all just conjecture of course though.

Rednaxela (talk)04:19, 17 January 2014

No proofs and only conjectures, but convincing enough.

I´ll try using multiple classifications generated at random in my next version and see what happens.

MN (talk)13:46, 17 January 2014
 
 
 
 

DC is more versatile than VCS. There is more room for improvement, but also more room to screw up. And WaveSim doesn't work with movement.

It is harder to unlock all the potential from DC in movement.

MN (talk)01:20, 14 January 2014