AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Optimal Gradient Checkpoint Search for Arbitrary Computation Graphs

Jianwei Feng, Dong Huang

CVPR 2021 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 17 citations (OpenAlex)

Check 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
Use the optimal checkpoint search instead of Chen's checkpoint searchGCP training · Vgg19 batch 64training memory (MB): 3957 -> 3565Improved
Use the optimal checkpoint search instead of Chen's checkpoint searchGCP training · Vgg16 batch 64training memory (MB): 3957 -> 3565Improved
Use the optimal checkpoint search instead of Chen's checkpoint searchGCP training · Vgg11 batch 64training memory (MB): 2781 -> 2781No clear effect
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · ResNet18 batch 256training memory (MB): 4069 -> 3677Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · ResNet34 batch 128training memory (MB): 2231 -> 1838Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · ResNet50 batch 64training memory (MB): 2714 -> 1973Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · ResNet101 batch 32training memory (MB): 2541 -> 1024Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · ResNet152 batch 16training memory (MB): 1464 -> 526Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · Densenet121 batch 32training memory (MB): 1629 -> 898Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · Densenet161 batch 16training memory (MB): 1432 -> 666Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · Densenet169 batch 32training memory (MB): 1774 -> 897Improved
Use the optimal checkpoint search instead of the best random checkpoint selectionGCP training · Densenet201 batch 16training memory (MB): 1242 -> 474Improved

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