Abstract
Decision forests can be thought of as a flexible optimization toolbox with many avenues to alter or recombine the underlying architectural components and improve recognition accuracy and efficiency. In this chapter, we present two fundamental approaches for re-architecting decision forests that yield higher prediction accuracy and shortened decision time.
The first is entanglement, i.e. using the learned tree structure and intermediate probabilities computed in nodes closer to the root to affect the training of other nodes deeper in the trees. Unlike more conventional classifiers which assume that all data points (even those neighboring in space or time) are IID, the entanglement approach learns semantic correlation in non IID data. To demonstrate, we build an entangled decision forest (EDF) that exploits spatial correlation in human anatomy by simultaneously labeling voxels in computed tomography (CT) scans into 12 anatomical structures.
The second contribution is the formulation of information gain as a function that is differentiable with respect to the parameters of the split node weak learner. This provides increased confidence and accuracy of maximum margin boundary localization and reduces classification time by using a few, shallow trees. We further extend the method to incorporate training label confidence, when available, into the information gain maximization. Due to bagging and random feature subset selection, we can retain decision forest virtues such as resiliency to overfitting. To demonstrate, we build a gradient ascent decision forest (GADF) that tracks visual objects in videos. For both approaches, superior accuracy and computational efficiency is shown in quantitative comparisons with state of the art algorithms.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amit Y, Geman D (1997) Shape quantization and recognition with randomized trees. Neural Comput 9(7)
Avidan S (2001) Support vector tracking. In: Proc IEEE conf computer vision and pattern recognition (CVPR), vol 1
Avidan S (2007) Ensemble tracking. IEEE Trans Pattern Anal Mach Intell 29(2)
Breiman L (2001) Random forests. Mach Learn 45(1)
Budvytis I, Badrinarayanan V, Cipolla R (2010) Label propagation in complex video sequences using semi-supervised learning. In: Proc British machine vision conference (BMVC)
Comaniciu D, Meer P (2002) Mean shift: a robust approach toward feature space analysis. IEEE Trans Pattern Anal Mach Intell 24(5)
Criminisi A, Shotton J, Bucciarelli S (2009) Decision forests with long-range spatial context for organ localization in CT volumes. In: MICCAI workshop on probabilistic models for medical image analysis (PMMIA)
Criminisi A, Shotton J, Robertson D, Konukoglu E (2010) Regression forests for efficient anatomy detection and localization in CT studies. In: MICCAI workshop on medical computer vision: recognition techniques and applications in medical imaging, Beijing. Springer, Berlin
Criminisi A, Shotton J, Konukoglu E (2012) Decision forests: a unified framework for classification, regression, density estimation, manifold learning and semi-supervised learning. Found Trends Comput Graph Vis 7(2–3)
Everingham M, van Gool L, Williams C, Winn J, Zisserman A (2010) The Pascal visual object classes (VOC) challenge 2010. Int J Comput Vis 88
Frank A, Asuncion A (2010) UCI machine learning repository
Gall J, Yao A, Razavi N, van Gool LJ, Lempitsky VS (2011) Hough forests for object detection, tracking, and action recognition. IEEE Trans Pattern Anal Mach Intell 33(11)
Geremia E, Menze B, Clatz O, Konukoglu E, Criminisi A, Ayache N (2010) Spatial decision forests for MS lesion segmentation in multi-channel MR images. In: Proc medical image computing and computer assisted intervention (MICCAI). Springer, Berlin
Geurts P, Ernst D, Wehenkel L (2006) Extremely randomized trees. Mach Learn 36(1)
Grabner H, Grabner M, Bischof H (2006) Real-time tracking via on-line boosting. In: Proc British machine vision conference (BMVC)
Heath D, Kasif S, Salzberg S (1993) Induction of oblique decision trees. J Artif Intell Res 2(2)
John GH (1995) Robust linear discriminant trees. In: Fifth intl workshop on artificial intelligence and statistics
Kalal Z, Matas J, Mikolajczyk K (2010) P-N learning: bootstrapping binary classifiers by structural constraints. In: Proc IEEE conf computer vision and pattern recognition (CVPR)
Lempitsky V, Verhoek M, Noble A, Blake A (2009) Random forest classification for automatic delineation of myocardium in real-time 3D echocardiography. In: Workshop on functional imaging and modelling of the heart (FIMH). Springer, Berlin
Menze B, Kelm BM, Splitthoff DN, Koethe U, Hamprecht FA (2011) On oblique random forests. In: Proc European conf on machine learning (ECML/PKDD)
Montillo A (2011) Context selective decision forests and their application to lung segmentation in CT images. In: MICCAI workshop on pulmonary image analysis
Montillo A, Shotton J, Winn J, Iglesias J, Metaxas D, Criminisi A (2011) Entangled decision forests and their application for semantic segmentation of CT images. In: Proc information processing in medical imaging (IPMI). Springer, Berlin
Murthy SK, Kasif S, Salzberg S (1994) A system for induction of oblique decision trees. arXiv:cs/9408103
Quinlan JR (1993) C4.5: programs for machine learning. Morgan Kaufmann, San Mateo
Saffari A, Leistner C, Santner J, Godec M, Bischoff H (2009) On-line random forests. In: ICCV workshop on on-line learning for computer vision
Shotton J, Johnson M, Cipolla R (2008) Semantic texton forests for image categorization and segmentation. In: Proc IEEE conf computer vision and pattern recognition (CVPR)
Shotton J, Winn JM, Rother C, Criminisi A (2009) TextonBoost for image understanding: multi-class object recognition and segmentation by jointly modeling texture, layout, and context. Int J Comput Vis 81(1)
Tu Z (2005) Probabilistic boosting-tree: learning discriminative models for classification, recognition, and clustering. In: Proc IEEE intl conf on computer vision (ICCV), Beijing, China, October 2005, vol 2
Tu Z, Bai X (2010) Auto-context and its application to high-level vision tasks and 3D brain image segmentation. IEEE Trans Pattern Anal Mach Intell 32(10)
UCI Machine Learning Repository. http://archive.ics.uci.edu/ml/datasets.html
Vedaldi A, Blaschko M, Zisserman A (2011) Learning equivariant structured output SVM regressors. In: Proc IEEE intl conf on computer vision (ICCV)
Yi Z, Criminisi A, Shotton J, Blake A (2009) Discriminative, semantic segmentation of brain tissue in MR images. In: Proc medical image computing and computer assisted intervention (MICCAI). Springer, Berlin
Yin P, Criminisi A, Winn J, Essa I (2007) Tree based classifiers for bilayer video segmentation. In: Proc IEEE conf computer vision and pattern recognition (CVPR)
Zheng Y, Georgescu B, Comaniciu D (2009) Marginal space learning for efficient detection of 2D/3D anatomical structures in medical images. In: Proc information processing in medical imaging (IPMI). Springer, Berlin
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag London
About this chapter
Cite this chapter
Montillo, A. et al. (2013). Entanglement and Differentiable Information Gain Maximization. In: Criminisi, A., Shotton, J. (eds) Decision Forests for Computer Vision and Medical Image Analysis. Advances in Computer Vision and Pattern Recognition. Springer, London. https://doi.org/10.1007/978-1-4471-4929-3_19
Download citation
DOI: https://doi.org/10.1007/978-1-4471-4929-3_19
Publisher Name: Springer, London
Print ISBN: 978-1-4471-4928-6
Online ISBN: 978-1-4471-4929-3
eBook Packages: Computer ScienceComputer Science (R0)