AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Zero Time Waste: Recycling Predictions in Early Exit Neural Networks

Maciej Wolczyk, Bartosz Wójcik, Klaudia Bałazy, Igor T. Podolak, Jacek Tabor, Marek Śmieja, Tomasz Trzcinski

NeurIPS 2021 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 33 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
Reuse previous internal-classifier outputs with cascade connections and geometric ensemblingEarly-exit ResNet-50 · ImageNet (25% compute)accuracy: 33.8 -> 34.9Improved
Reuse previous internal-classifier outputs with cascade connections and geometric ensemblingEarly-exit ResNet-50 · ImageNet (50% compute)accuracy: 53.8 -> 54.9Improved
Reuse previous internal-classifier outputs with cascade connections and geometric ensemblingEarly-exit ResNet-50 · OCT2017 (25% compute)accuracy: 81.5 -> 89.4Improved
Use both cascade connections and geometric ensembling rather than either technique alone (ablation)ZTW · CIFAR-100 on ResNet-56 and VGG-16accuracyImproved
Use geometric rather than additive ensembling (ablation)ZTW with cascade connections · ResNet-56 on CIFAR-10, CIFAR-100, and Tiny ImageNetaccuracyImproved
Stop gradients through cascade connections (ablation)ZTW · ResNet-56 on CIFAR-10, CIFAR-100, and Tiny ImageNetaccuracyMixed
Apply ZTW to binary rather than 10-class classification (ablation)Early-exit models · CIFAR-2accuracyMixed

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