Open Set Recognition for Random Forest

TL;DR

This paper introduces a novel method to enable random forest classifiers to recognize unknown classes in open-set scenarios by integrating distance metric learning and distance-based recognition, validated on synthetic and real datasets.

Contribution

It proposes a new approach that extends random forest capabilities to open-set recognition using distance metrics, which was not previously available.

Findings

Outperforms existing open-set recognition methods

Effective on both synthetic and real-world datasets

Enhances random forest applicability in open-set scenarios

Abstract

In many real-world classification or recognition tasks, it is often difficult to collect training examples that exhaust all possible classes due to, for example, incomplete knowledge during training or ever changing regimes. Therefore, samples from unknown/novel classes may be encountered in testing/deployment. In such scenarios, the classifiers should be able to i) perform classification on known classes, and at the same time, ii) identify samples from unknown classes. This is known as open-set recognition. Although random forest has been an extremely successful framework as a general-purpose classification (and regression) method, in practice, it usually operates under the closed-set assumption and is not able to identify samples from new classes when run out of the box. In this work, we propose a novel approach to enabling open-set recognition capability for random forest classifiers by incorporating distance metric learning and distance-based open-set recognition. The proposed method is validated on both synthetic and real-world datasets. The experimental results indicate that the proposed approach outperforms state-of-the-art distance-based open-set recognition methods.