Alternating Decision Forests

By Samuel Schulter, Paul Wohlhart, Christian Leistner, Amir Saffari, Peter M. Roth, Horst Bischof. 2013

73 citations recorded by the source corpus.

This paper introduces a novel classification method termed Alternating Decision Forests (ADFs), which formulates the training of Random Forests explicitly as a global loss minimization problem. During training, the losses are minimized via keeping an adaptive weight distribution over the training samples, similar to Boosting methods. In order to keep the method as flexible and general as possible, we adopt the principle of employing gradient descent in function space, which allows to minimize arbitrary losses. Contrary to Boosted Trees, in our method the loss minimization is an inherent part of the tree growing process, thus allowing to keep the benefits of common Random Forests, such as, parallel processing. We derive the new classifier and give a discussion and evaluation on standard machine learning data sets. Furthermore, we show how ADFs can be easily integrated into an object detection application. Compared to both, standard Random Forests and Boosted Trees, ADFs give better performance in our experiments, while yielding more compact models in terms of tree depth.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, background): “This is similar in spirit to [35], where a weight distribution is utilized to improve the learning process of random forest.”
End-to-End Learning of Decision Trees and Forests (2019, background): “Various works are now exploring the relation between both classiﬁcation approaches (Ioannou et al. 2016; Wang et al. 2017), such as the Deep Neural Decision Forests (DNDFs) (Kontschieder et al. 2015).”
Adaptive Neural Trees (2018, unknown): “21 – – 200 Alternating Decision Forest [33] 2.”
Deep Neural Decision Forests (2015, methodology): “We used the 5 datasets in [30] to compare the performance of sNDFs to that of Alternating Decision Forests (ADF).”
Features in Concert: Discriminative Feature Selection meets Unsupervised Clustering (2014, background): “There is also work [43] on combining decision forests with ideas from boosting, in order to obtain a weighted average of trees that better fits the training data.”
Decision Forests, Convolutional Networks and the Models in-Between (2016, unknown): “[11, 19, 25]).”
Alternating Regression Forests for Object Detection and Pose Estimation (2013, methodology): “The recently proposed ADFs [19] tackle this issue by reformulating the training phase of RFs as a stage-wise classifier.”