AutoEncoder by Forest

TL;DR

This paper introduces EncoderForest, a novel tree ensemble auto-encoder that achieves lower reconstruction errors and faster training compared to traditional neural network autoencoders, with added benefits of reusability and damage tolerance.

Contribution

It presents the first tree ensemble based auto-encoder, enabling backward reconstruction through decision path equivalence classes in forests.

Findings

Lower reconstruction error than DNN autoencoders

Faster training speed

Reusability and damage tolerance of the model

Abstract

Auto-encoding is an important task which is typically realized by deep neural networks (DNNs) such as convolutional neural networks (CNN). In this paper, we propose EncoderForest (abbrv. eForest), the first tree ensemble based auto-encoder. We present a procedure for enabling forests to do backward reconstruction by utilizing the equivalent classes defined by decision paths of the trees, and demonstrate its usage in both supervised and unsupervised setting. Experiments show that, compared with DNN autoencoders, eForest is able to obtain lower reconstruction error with fast training speed, while the model itself is reusable and damage-tolerable.