version 3.1.0.0 (5.97 KB) by
Kota Hara

Boosted Binary Regression Trees is a powerful regression method which can handle vector targets.

Boosted Binary Regression Trees (BBRT) is a powerful regression method proposed in [1]. BBRT combines binary regression trees [3] using a gradient boosting technique.

There are several variants proposed in [1]. In [1], it is assumed that the target is a scalar value. However, it is trivial to extend the method to vector target cases by proper modifications.

This code is based on "LS_Boost" described in [1] but it can also handle vector target cases. In other words, you do not need to train an independent regressor for each target dimension, unlike Support Vector Regression.

The detail of the algorithm this code implements can be found in [2].

[1] J. H. Friedman. Greedy Function Approximation: a Gradient Boosting Machine. Annals of Statistics, 2001.

[2] Kota Hara and Rama Chellappa, Computationally Efficient Regression on a Dependency Graph for Human Pose Estimation, CVPR 2013.

[3] Breiman, Leo; Friedman, J. H., Olshen, R. A., & Stone, C. J. (1984). Classification and regression trees. Monterey, CA: Wadsworth & Brooks/Cole Advanced Books & Software. ISBN 978-0-412-04841-8.

Kota Hara (2021). Boosted Binary Regression Trees (https://www.mathworks.com/matlabcentral/fileexchange/42130-boosted-binary-regression-trees), MATLAB Central File Exchange. Retrieved .

Created with
R2014a

Compatible with any release

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!Create scripts with code, output, and formatted text in a single executable document.

smoothie tncan you please give an exemple of application of these functions (in one code)? thanks

yuan shuaibiLai WeiHello, how can I get the importance of each element?Thank you

whichDuo LiHassan mohamedHello, I tried the code but it gives me only one number as an output without varargin sentence or error message of : Attempts to execute varargin.m file to bath

c:\ program files \ MATALB\ .....\lang\varargin.m

please give advice.

Tom ClarkRemove the include of tchar.h. It contains no dependencies and prevents compilation on non-windows platforms.

KotaI released the code for new regression tree forests method described in "Growing Regression Forests by Classification: Applications to Object Pose Estimation. The European Conference on Computer Vision (ECCV), 2014." The method outperforms various regression methods such as traditional binary regression tree forests as well as Support Vector Regression and Kernel Partial Least Squares Regression. The code is available from here. It is written in matlab.

http://www.kotahara.com/download-k-clusters-regression-forest.html

UTAHi, would you please add some date for code running? ths

Kota HaraHi Hossein,

I think the implementation is correct, though it's not the most efficient. The update procedures in this code can be proved to be equivalent to those proposed in "Updating Mean and Variance Estimates: An Improved Method." You can also find it in wikipedia. (en.wikipedia.org/wiki/Algorithms_for_calculating_variance)

The update procedures in the above method are:

SSE_{N} = SSE_{N-1} + (x_N - \bar x_{N-1})(x_N - \bar x_N)

The second term of RHS is

(x_N - \bar x_N + \bar x_N - \bar x_{N-1} ) ( x_N - \bar x_N ) = ( x_N - \bar x_N )^2 + (x_N - \bar x_N )(\bar x_N - \bar x_{N-1} )

The procedure in my code is

SSE_{N} = SSE_{N-1} + (\bar x_N - \bar x_{N-1})^2 * (N-1) + ( x_N - \bar x_N )^2

So now we want to show

(x_N - \bar x_N )(\bar x_N - \bar x_{N-1} ) = (\bar x_N - \bar x_{N-1})^2 * (N-1)

First divide both side by (\bar x_N - \bar x_{N-1}). Now we need to show (\bar x_N - \bar x_{N-1})(N-1) = x_N - \bar x_N

We can show this using the update procedure for mean (i.e., \bar x_N = \bar x_{N-1} + (x_N - \bar x_{N-1}) / N

The proof for sseRight should be done similarly.

As I said, this is not as efficient as the original procedure, so I will consider modifying the code.

Thanks.

HosseinHi Kota,

Are you sure that the implementation of findBestSplit is correct? I have doubts about the SSE update. I mean the lines 129 and 130

for( int k=0; k<targetDim; k++ ){

sseLeft = sseLeft + ( aveLeft[k] - aveLeftPre[k] ) * ( aveLeft[k] - aveLeftPre[k] ) * ( sizeLeft - 1 ) + ( target(sortByValue[j].idx,k) - aveLeft[k] ) * ( target(sortByValue[j].idx,k) - aveLeft[k] );

sseRight = sseRight - ( aveRightPre[k] - aveRight[k] ) * ( aveRightPre[k] - aveRight[k] ) * sizeRight - ( target(sortByValue[j].idx,k) - aveRightPre[k] ) * ( target(sortByValue[j].idx,k) - aveRightPre[k] );

}

I understand the procedure, but I think the update rule is not correct.

Thanks

HosseinI could compile on unix by removing #include <tchar.h> and 2 minor changes.

HosseinHow can I compile the cpp file in unix? Thank you.

Kota HaraPlease feel free to drop a comment if you have suggestions, find bugs or anything.