AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Lottery Tickets on a Data Diet: Finding Initializations with Sparse Trainable Networks

Mansheej Paul, Brett W Larsen, Surya Ganguli, Jonathan Frankle, Gintare Karolina Dziugaite

NeurIPS 2022 · Architectures and efficiency › Pruning and efficient tuning · approach: Pruning at initialization · 4 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
increase IMP rewind step from 0 to t*IMP sparse networks · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)final test accuracy at high sparsityImproved
pre-train on small random subset for t*/2 instead of all data for t*/2IMP initialization · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)final test accuracyImproved
pre-train on easiest (lowest EL2N) examples for t*/2IMP initialization · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)final test accuracyImproved
pre-train on hardest (highest EL2N) examples for t*/2IMP initialization · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)final test accuracyHurt
add 10% randomized labels during pre-training: easiest-data subset vs random subset or all dataIMP initialization · CIFAR-10final test accuracyImproved
increase randomized labels to 50% during pre-training with all data or random subsetIMP initialization · CIFAR-10final test accuracyHurt
pre-train on lowest-train-loss-barrier subset (ablation)IMP initialization · CIFAR-100/CINIC-10final test accuracyMixed
pre-train on easy data (hypothesis: acts as gradient clipping) (ablation)dense network pre-training · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)minibatch gradient normHurt
use mean per-example train loss barrier as predictor instead of pre-train accuracydense initialization · CIFAR-10/CIFAR-100/CINIC-10 (ResNet-20/32/56)correlation with final test accuracyImproved

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