Deep Neural Decision Forests

By Peter Kontschieder, Madalina Fiterau, Antonio Criminisi, Samuel Rota Bulò. 2015

557 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Plant Science, Model Architectures

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

In Plant Science

In Model Architectures

Citations in the corpus

Dynamic Neural Networks: A Survey (2021, background): “a) Soft decision tree (SDT) [98], [99], [100] adopts neural units as routing nodes (blue nodes in Fig.”
Neural Prototype Trees for Interpretable Fine-grained Image Recognition (2020, background): “In contrast to traditional decision trees, where an internal node routes sample z either right or left, our node n ∈ N is soft and routes z to both children, each with a fuzzy weight within [0, 1], giving it a probabilistic interpretation [15, 23, 29, 47].”
MurTree: Optimal Decision Trees via Dynamic Programming and Search (2020, methodology): “…methods (Blanc et al. (2020)), a ﬁne-grained computational complexity study (Ordyniak and Szei-der (2021)), neural networks for decision trees (Kontschieder et al. (2015); Tanno et al. (2019)), randomised trees (Blanquero et al. (2020)), end-to-end learning of decision trees (Hehn et al.…”
Deep Neural Decision Trees (2018, methodology): “E.g., in Kontschieder et al. (2015), they use a less flexible tree where the structure is fixed while the node split is learned.”
XOC: Explainable Observer-Classifier for Explainable Binary Decisions (2019, methodology): “The prior work closest to ours is the Deep Neural Decision Forest (Kontschieder et al., 2016), which ﬁrst uses a CNN to determine the routing probabilities on each node and then combines nodes to an ensemble of decision trees that jointly make the prediction.”
Tree-CNN: A hierarchical Deep Convolutional Neural Network for incremental learning (2018, background): “Deep Neural Decision Forests [24] uniﬁed decision trees and deep CNN’s to build a hierarchical classiﬁer.”
Deep Facial Expression Recognition: A Survey (2018, result): “investigated the adaptation of deep neural forests (NFs) [95], which replaces the softmax loss with NFs and achieved competitive results for FER.”
HydraNets: Specialized Dynamic Architectures for Efficient Inference (2018, methodology): “The hyper parameters k, nb and the point to branch were chosen empirically by searching over nb = [5, 10, 20, 25], k = [3, 4, 5] and branching at block 2 or block 3.”
iPrivacy: Image Privacy Protection by Identifying Sensitive Objects via Deep Multi-Task Learning (2017, background): “Index Terms— Image sharing, privacy setting recommendation, object-privacy alignment, image privacy protection, privacysensitive object classes, deep multi-task learning, tree classifier for hierarchical object detection.”
The Tree Ensemble Layer: Differentiability meets Conditional Computation (2020, methodology): “For example, Kontschieder et al. (2015) equipped soft trees with neural representations and used alternating minimization to learn the feature representations and the”
SegNBDT: Visual Decision Rules for Segmentation (2020, unknown): “[12] analyze a prior deep learning decision tree (dNDF) [10] for MNIST and CIFAR-10.”
Weighted Oblique Decision Trees (2019, methodology): “Generally, oblique decision trees simplify tree structure and achieve better performance (Breiman et al. 1984; Heath, Kasif, and Salzberg 1993; Brodley and Utgoff 1995; Loh and Shih 1997; Amasyali and Ersoy 2008; Robertson, Price, and Reale 2013; Kontschieder et al. 2015).”
Decision Forests, Convolutional Networks and the Models in-Between (2016, unknown): “[11, 19, 25]).”
Visualizing the decision-making process in deep neural decision forest (2019, background): “Deep neural decision forest (NDF) [4] and its later regression version [6] formulated a probabilistic routing framework for decision trees.”
TabNet: Attentive Interpretable Tabular Learning (2019, background): “Yet, conventional DNNs, DTs, and even their hybrid variant of deep neural DTs (Yang, Morillo, and Hospedales 2018) severely suffer from the imbalanced data and cannot learn the required sorting and ranking operations (Yang, Morillo, and Hospedales 2018).”
Deciding How to Decide: Dynamic Routing in Artificial Neural Networks (2017, methodology): “More recently, Kontschieder et al. (2015) performed joint optimization of ANN and decision tree parameters, and Bulo & Kontschieder (2014) used randomized multi-layer networks to compute decision tree split functions.”
A-ViT: Adaptive Tokens for Efficient Vision Transformer (2021, background): “One way is to change the architecture of the model to support adaptive computations [4,14,15, 18, 25, 27, 30, 37, 42, 51, 54].”