Denoising random forests
This addresses a specific issue for users of random forests in noisy data environments, but it is an incremental improvement combining existing techniques.
The paper tackles the problem of random forests being vulnerable to noise in test samples, which degrades estimation performance, by proposing denoising random forests that use denoising autoencoders to identify and correct incorrect node decisions, resulting in improved robustness.
This paper proposes a novel type of random forests called a denoising random forests that are robust against noises contained in test samples. Such noise-corrupted samples cause serious damage to the estimation performances of random forests, since unexpected child nodes are often selected and the leaf nodes that the input sample reaches are sometimes far from those for a clean sample. Our main idea for tackling this problem originates from a binary indicator vector that encodes a traversal path of a sample in the forest. Our proposed method effectively employs this vector by introducing denoising autoencoders into random forests. A denoising autoencoder can be trained with indicator vectors produced from clean and noisy input samples, and non-leaf nodes where incorrect decisions are made can be identified by comparing the input and output of the trained denoising autoencoder. Multiple traversal paths with respect to the nodes with incorrect decisions caused by the noises can then be considered for the estimation.