"Optimal" Predictor, help me understanding adaptive covariance estimation 
"Optimal" Predictor, help me understanding adaptive covariance estimation 
Nov 15 2006, 21:16
Post
#1


Group: Members Posts: 208 Joined: 12March 04 From: Germany Member No.: 12686 
As some of you may have noticed i'm also working on a losslessaudiocodec.
The core prediction uses a jointchannel cholesky decomposition every ksamples. The decomposition is applied on a covariance matrix which is updated every sample in the following way CODE for (int i=0;i<=maxorder;i++) { for (int j=i;j<=maxorder;j++) covar[i][j] = decay * covar[i][j] + history[i] * history[j] } You can see there's a learning factor 'decay' involved. The problem is that for some samples olddata seems to be more important than recent data leading to a large encodingcost if you use a relative high factor that works good on most samples. I've tried to exploit some statistical properties as prediction gain or variance of the input to estimate a optimal learning factor but this seems to be more complicated than i first thought. Does anybody know of a good way to estimate this property? any help is appreciated pest edit: typo in the codebox This post has been edited by pest: Nov 16 2006, 14:13 


Nov 15 2006, 22:21
Post
#2


Group: Developer Posts: 1318 Joined: 20March 04 From: Göttingen (DE) Member No.: 12875 
Looks like you want to do backward adaptive prediction. Are you sure you want a decoder to perform the same calculations (cholesky every k samples)? This is pretty time consuming, isn't it? Also, if you want your compressed files to be machineindependent you need to make sure, that the calculations you perform give the exact same results on every machine. (This is not the case if you rely on an FPU for these calculations.)
Backward adaptive prediction is not my specialty, sorry. Perhaps a sliding window approach helps. This post has been edited by SebastianG: Nov 15 2006, 22:25 


LoFi Version  Time is now: 23rd November 2014  17:28 