Deep Forest with Hashing Screening and Window Screening
This work addresses efficiency and performance issues in deep forest models for machine learning applications, but it is incremental as it builds on existing gcForest methods.
The paper tackles the problem of redundant feature vectors in gcForest's multi-grained scanning, which increases time cost, by introducing HW-Forest with hashing and window screening strategies, resulting in higher accuracy and reduced time and memory consumption.
As a novel deep learning model, gcForest has been widely used in various applications. However, the current multi-grained scanning of gcForest produces many redundant feature vectors, and this increases the time cost of the model. To screen out redundant feature vectors, we introduce a hashing screening mechanism for multi-grained scanning and propose a model called HW-Forest which adopts two strategies, hashing screening and window screening. HW-Forest employs perceptual hashing algorithm to calculate the similarity between feature vectors in hashing screening strategy, which is used to remove the redundant feature vectors produced by multi-grained scanning and can significantly decrease the time cost and memory consumption. Furthermore, we adopt a self-adaptive instance screening strategy to improve the performance of our approach, called window screening, which can achieve higher accuracy without hyperparameter tuning on different datasets. Our experimental results show that HW-Forest has higher accuracy than other models, and the time cost is also reduced.