AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Training Decision Trees as Replacement for Convolution Layers

Wolfgang Fuhl, Gjergji Kasneci, Wolfgang Rosenstiel, Enkeljda Kasneci

AAAI 2020 · Learning theory and algorithms › Prediction and decision losses · approach: Optimal decision tree learning · 44 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace convolution layers with TI2 decision-tree layers (ablation)LeNet-5 · MNIST testclassification accuracy: 99.37 -> 99.23Hurt
Use TI3 index pattern with inception instead of convolution layers (ablation)LeNet-5 · MNIST testclassification accuracy: 99.37 -> 99.48Improved
Replace convolution layers with TI1 decision-tree layersResNet-34 · CIFAR10 testclassification accuracy: 91.12 -> 92.20Improved
Replace convolution layers with TI1 decision-tree layersResNet-16 · 300W commonNME: 3.69 -> 3.64Improved
Replace convolution layers with TI1 decision-tree layersResNet-16 · 300W challengingNME: 6.76 -> 6.81Hurt
Replace convolution layers with TI1 decision-tree layersResNet-16 · 300W fullNME: 4.29 -> 4.26Improved
Replace convolution layers with TI2 decision-tree layers (ablation)LeNet-5 · MNIST test inference, 1 CPU coreruntime: 0.83ms -> 0.18msImproved
Use TI3 rather than TI2 index pattern (ablation)LeNet-5 · MNIST test inference, 1 CPU coreruntime: 0.18ms -> 0.36msHurt
Replace convolution layers with TI1 decision-tree layersResNet-34 · CIFAR10 test inference, GPU 1050tiruntime: 18.1ms -> 6.77msImproved
Replace convolution layers with TI1 decision-tree layersResNet-16 · 300W test inference, GPU 1050tiruntime: 10.76ms -> 4.64msImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy