AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

A Graph Theoretic Framework of Recomputation Algorithms for Memory-Efficient Backpropagation

Mitsuru Kusumoto, Takuya Inoue, Gentaro Watanabe, Takuya Akiba, Masanori Koyama

NeurIPS 2019 · Graph learning › Graph algorithms and clustering · approach: Neural combinatorial optimization on graphs · 22 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
Use graph-theoretic recomputation with a memory-centric strategy instead of Chen's algorithmApproxDP + MC · PSPNet, U-Net, ResNet50, ResNet152, VGG19, DenseNet161, GoogLeNetpeak memory consumption: PSPNet: 4.0 GB -> 2.7 GB; U-Net: 7.4 GB -> 4.7 GB; ResNet50: 3.7 GB -> 3.4 GB; ResNet152: 2.4 GB -> 2.3 GB; VGG19: 4.7 GImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · PSPNetpeak memory consumption: 3.2 GB -> 2.7 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · U-Netpeak memory consumption: 6.8 GB -> 5.0 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · ResNet50peak memory consumption: 4.7 GB -> 3.4 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · ResNet152peak memory consumption: 2.8 GB -> 2.3 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · VGG19peak memory consumption: 5.5 GB -> 4.5 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · DenseNet161peak memory consumption: 1.9 GB -> 1.6 GBImproved
Add liveness analysis to ApproxDP + MC (ablation)ApproxDP + MC · GoogLeNetpeak memory consumption: 6.0 GB -> 5.2 GBImproved
Use memory-centric rather than time-centric strategy (ablation)ApproxDP · PSPNet, U-Net, ResNet50, ResNet152, VGG19, DenseNet161, GoogLeNetpeak memory consumption: PSPNet: 3.1 GB -> 2.7 GB; U-Net: 6.7 GB -> 5.0 GB; ResNet50: 4.4 GB -> 3.4 GB; ResNet152: 2.5 GB -> 2.3 GB; VGG19: 5.5 GMixed
Use approximate DP rather than exact DP (ablation)DP recomputation strategy · benchmark networkspeak memory consumptionNo clear effect
Use approximate DP rather than exact DP (ablation)DP recomputation strategy · benchmark networksDP algorithm runtime: within 1 sec versus more than 80 secsImproved
Use ApproxDP + MC rather than Chen's algorithm without liveness analysisrecomputation strategy · PSPNetpeak memory consumption: 7.6 GB -> 3.2 GBImproved

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