mlsanityData to 7 Oct 2026

Neural Networks are Decision Trees

arXiv 2210.05189 ↗First posted 11 Oct 2022cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

In this manuscript, we show that any neural network with any activation function can be represented as a decision tree. The representation is equivalence and not an approximation, thus keeping the accuracy of the neural network exactly as is. We believe that this work provides better understanding of neural networks and paves the way to tackle their black-box nature. We share equivalent trees of some neural networks and show that besides providing interpretability, tree representation can also achieve some computational advantages for small networks. The analysis holds both for fully connected and convolutional networks, which may or may not also include skip connections and/or normalizations.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 22 of its 31 references that are in the map. First posted on arXiv 11 Oct 2022.

Builds on

22 of 31 references.

References · 22

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2015Learning Deep Features for Discriminative Localizationcited by 2,430
  3. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  4. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  5. 2016YOLO9000: Better, Faster, Strongercited by 1,429
  6. 2017Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networkscited by 936
  7. 2016Top-down Neural Attention by Excitation Backpropcited by 318
  8. 2017Distilling a Neural Network Into a Soft Decision Treecited by 230
  9. 2020Eigen-CAM: Class Activation Map using Principal Componentscited by 188
  10. 2017Convolutional Networks with Adaptive Inference Graphscited by 147
  11. 2018Tropical Geometry of Deep Neural Networkscited by 92
  12. 2020Use HiResCAM instead of Grad-CAM for faithful explanations of convolutional neural networkscited by 79
10 more
  1. 2018Deep Neural Decision Treescited by 72
  2. 2017Beyond Sparsity: Tree Regularization of Deep Models for Interpretabilitycited by 68
  3. 2018Deep Feature Factorization For Concept Discoverycited by 64
  4. 2016Network of Experts for Large-Scale Image Categorizationcited by 54
  5. 2018Mad Max: Affine Spline Insights into Deep Learningcited by 54
  6. 2020NBDT: Neural-Backed Decision Treescited by 46
  7. 2017Deciding How to Decide: Dynamic Routing in Artificial Neural Networkscited by 33
  8. 2020Unwrapping The Black Box of Deep ReLU Networks: Interpretability, Diagnostics, and Simplificationcited by 32
  9. 2015Blockout: Dynamic Model Selection for Hierarchical Deep Networkscited by 11
  10. 2020Towards Interpretable ANNs: An Exact Transformation to Multi-Class Multivariate Decision Treescited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 8; 2024: 4; 2025: 2; 2026: 2.

16 papers.

Sort:

By area:

  1. 2024Interpretable Concept Bottlenecks to Align Reinforcement Learning Agentscited by 23
  2. 2023SoK: A Systematic Evaluation of Backdoor Trigger Characteristics in Image Classificationcited by 6
  3. 2023Unwrapping All ReLU Networkscited by 4
  4. 2023Establishing Trustworthiness: Rethinking Tasks and Model Evaluationcited by 4
  5. 2024Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignmentcited by 3
  6. 2025Prophecy: Inferring Formal Properties from Neuron Activationscited by 2
  7. 2024The Topos of Transformer Networkscited by 1
  8. 2023Myths and Legends in High-Performance Computingcited by 0
  9. 2023Efficient Multi-stage Inference on Tabular Datacited by 0
  10. 2023LEURN: Learning Explainable Univariate Rules with Neural Networkscited by 0
  11. 2023Model of models -- Part 1cited by 0
  12. 2023High-Level Parallelism and Nested Features for Dynamic Inference Cost and Top-Down Attentioncited by 0
4 more
  1. 2024Extracting PAC Decision Trees from Black Box Binary Classifiers: The Gender Bias Case Study on BERT-based Language Modelscited by 0
  2. 2025Even Faster Hyperbolic Random Forests: A Beltrami-Klein Wrapper Approachcited by 0
  3. 2026Poisson Hyperplane Processes with Rectified Linear Unitscited by 0
  4. 2026Late Breaking Results: Conversion of Neural Networks into Logic Flows for Edge Computingcited by 0