Deep forest

By Zhi-Hua Zhou, Ji Feng. International Joint Conference on Artificial Intelligence, 2017

1,144 citations recorded by the source corpus.

Abstract Current deep-learning models are mostly built upon neural networks, i.e. multiple layers of parameterized differentiable non-linear modules that can be trained by backpropagation. In this paper, we explore the possibility of building deep models based on non-differentiable modules such as decision trees. After a discussion about the mystery behind deep neural networks, particularly by contrasting them with shallow neural networks and traditional machine-learning techniques such as decision trees and boosting machines, we conjecture that the success of deep neural networks owes much to three characteristics, i.e. layer-by-layer processing, in-model feature transformation and sufficient model complexity. On one hand, our conjecture may offer inspiration for theoretical understanding of deep learning; on the other hand, to verify the conjecture, we propose an approach that generates deep forest holding these characteristics. This is a decision-tree ensemble approach, with fewer hyper-parameters than deep neural networks, and its model complexity can be automatically determined in a data-dependent way. Experiments show that its performance is quite robust to hyper-parameter settings, such that in most cases, even across different data from different domains, it is able to achieve excellent performance by using the same default setting. This study opens the door to deep learning based on non-differentiable modules without gradient-based adjustment, and exhibits the possibility of constructing deep models without backpropagation.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7ea35b35392c6ef5738635cec7d17b24fe3e4f04

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

A survey on ensemble learning (2019, methodology): “Some researches focus on relaxing limitations of existing methods, such as Zhou and Feng [56] presented the multi-grained cascade forest method to rival deep neural networks (DNN) by achieving comparable performances, in which a deep forest ensemble with a cascade structure was generated for represe”
Neural Oblivious Decision Ensembles for Deep Learning on Tabular Data (2019, methodology): “Another line of work (Miller et al., 2017; Zhou & Feng, 2017; Feng et al., 2018) promotes the construction of multi-layer architectures from non-differentiable blocks, such as random forests or GBDT ensembles.”
The Emerging Trends of Multi-Label Learning (2020, methodology): “As an alternative of DNN, deep forest [161] is a recent deep learning framework based on tree model ensembles,”
One Venue, Two Conferences: The Separation of Chinese and American Citation Networks (2022, background): “Deep Forests [20], which were developed at Nanjing University, garnered over 600 citations that are mostly Chinese.”
Ensemble deep learning: A review (2021, methodology): “Following the stacking based approach, Deep convex net (DCN) [79], traditional methods like random forest [5, 97], support vector machines [94, 95, 96] have been extended to deep learning architectures which resulted in improved performance.”
Epileptic Seizure Detection Based on Variational Mode Decomposition and Deep Forest Using EEG Signals (2022, methodology): “Deep forest (DF) is a new deep ensemble method proposed by Zhou et al., which combines deep neural networks and decision trees [27].”