Back

Neural-Backed Decision Trees

Alvin Wan, Lisa Dunlap*, Daniel Ho*, Jihan Yin, Scott Lee, Henry Jin, Suzanne Petryk, Sarah Adel Bargal, Joseph E. Gonzalez

Our models, termed Neural-Backed Decision Trees, are as accurate as state-of-the-art neural networks on image classification and as interpretable as decision trees.

Try the demo

Colab

Provide our classification model with an image of your choice, or pick one of our suggested images. Unlike a run-of-the-mill neural network, our NBDT returns sequential decisions leading up to a prediction.

Decision 1

Animal

97.2% probability

Decision 2

Chordate

97.2% probability

Decision 3

Carnivore

97.2% probability

Prediction

Dog

97.2% probability

Authors

Alvin Wan, Lisa Dunlap*, Daniel Ho*, Jihan Yin, Scott Lee, Henry Jin, Suzanne Petryk, Sarah Adel Bargal, Joseph E. Gonzalez

*denotes equal contribution

Affiliations

University of California, Berkeley

Boston University

Published (Preprint)

April 1, 2020

Abstract

Machine learning applications such as finance and medicine demand accurate and justifiable predictions, barring most deep learning methods from use. In response, previous work combines decision trees with deep learning, creating a dilemma:

  1. sacrifice interpretability to maintain accuracy OR
  2. sacrifice accuracy to maintain interpretability.

We forgo this dilemma by proposing Neural-Backed Decision Trees (NBDTs), modified hierarchical classifiers that use trees constructed in weight-space. Our NBDTs achieve (1) interpretability and (2) neural network accuracy: We preserve interpretable properties -- e.g., leaf purity and a non-ensembled model -- and demonstrate interpretability of model predictions both qualitatively and quantitatively. Furthermore, NBDTs match state-of-the-art neural networks on CIFAR10, CIFAR100, TinyImageNet, and ImageNet to within 1-2%. This yields state-of-the-art interpretable models on ImageNet, with NBDTs besting all decision-tree-based methods by ~14% to attain 75.30% top-1 accuracy. Code and pretrained NBDTs are on Github.

Takeaways

Our work culminates in three key contributions that you can takeaway for future research:

Getting Started

Installation is just one line.

pip install nbdt

Run on any image of your choosing.

nbdt https://bit.ly/3eiuCId