AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Low-rank lottery tickets: finding efficient low-rank neural networks via matrix differential equations

Steffen Schotthöfer, Emanuele Zangrando, Jonas Kusch, Gianluca Ceruti, Francesco Tudisco

NeurIPS 2022 · Architectures and efficiency › Pruning and efficient tuning · approach: Lottery ticket discovery · 5 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
Train LeNet5 using adaptive low-rank factors at τ=0.11LeNet5 · MNISTtest accuracy: 99.2% -> 98.0%Hurt
Increase the adaptive truncation threshold from τ=0.11 to τ=0.3 (ablation)DLRT on LeNet5 · MNISTevaluation parameter compression: 88.86% -> 96.4%Improved
Increase the adaptive truncation threshold from τ=0.11 to τ=0.3 (ablation)DLRT on LeNet5 · MNISTtest accuracy: 98.0% -> 95.3%Hurt
Use DLRT rather than LeNet5 full-rank SGD trainingLeNet5 · MNISTevaluation parameter compression: 0% -> 96.4%Improved
Use DLRT rather than full-rank trainingResNet-50 · ImageNet1ktop-5 test accuracy difference to baseline: -0.56Mixed
Use DLRT rather than full-rank trainingVGG16 · ImageNet1ktop-5 test accuracy difference to baseline: -2.19Hurt
Use DLRT rather than full-rank trainingVGG16 · CIFAR-10test accuracy difference to baseline: -1.89Hurt
Use DLRT rather than full-rank trainingAlexNet · CIFAR-10test accuracy difference to baseline: -1.79Hurt
Truncate full-rank weight SVD without low-rank retraining (ablation)5-layer 784-neuron network · MNISTtest accuracy: 98.63% -> 9.91%Hurt
Add low-rank retraining after truncating the SVD at rank 100 (ablation)5-layer 784-neuron network · MNISTtest accuracy: 9.91% -> 98.16%Improved
Use fixed low-rank factors at ranks [5,5,5,5,10] instead of full-rank weights (ablation)5-layer 5120-neuron network · MNISTbatch training time: 0.320 -> 0.071Improved
Use fixed low-rank factors at ranks [5,5,5,5,10] instead of full-rank weights (ablation)5-layer 5120-neuron network · MNISTdataset prediction time: 1.2476 -> 0.2472Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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