We present the first tree-based regressor whose convergence rate depends only on the intrinsic dimension of the data, namely its Assouad dimension. The regressor uses the RPtree partitioning procedure, a simple randomized variant of k-d trees.
|Original language||English (US)|
|State||Published - 2009|
|Event||22nd Conference on Learning Theory, COLT 2009 - Montreal, QC, Canada|
Duration: Jun 18 2009 → Jun 21 2009
|Other||22nd Conference on Learning Theory, COLT 2009|
|Period||6/18/09 → 6/21/09|
All Science Journal Classification (ASJC) codes