AI Paper Atlas

Research map / Graph learning / GNN architectures

SyncTREE: Fast Timing Analysis for Integrated Circuit Design through a Physics-informed Tree-based Graph Neural Network

Yuting Hu, Jiajie Li, Florian Klemme, Gi-Joon Nam, Tengfei Ma, Hussam Amrouch, Jinjun Xiong

NeurIPS 2023 · Graph learning › GNN architectures · approach: Equivariant networks for molecules · 0 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 with TContrast loss instead of default L1 loss (ablation)SyncTREE (32 layers) · Synthetic datasetdelay MAE (ps): 3.424 -> 3.354Improved
train with TContrast loss instead of default L1 loss (ablation)SyncTREE (32 layers) · RISC-V datasetdelay MAE (ps): 0.0106 -> 0.0098Improved
train with TContrast loss instead of default L1 loss (ablation)SyncTREE (32 layers) · Synthetic datasetslew MAE (ps): 12.864 -> 12.608Improved
train with TContrast loss instead of default L1 loss (ablation)SyncTREE (4 layers) · RISC-V datasetslew MAE (ps): 0.0410 -> 0.0406Improved
replace L1 loss with Circle loss (ablation)SyncTREE (16 layers) · Synthetic datasetdelay MAE (ps): 3.775 -> 3.912Hurt
replace L1 loss with Circle loss (ablation)SyncTREE (8 layers) · RISC-V datasetdelay MAE (ps): 0.0195 -> 0.0223Hurt
replace L1 loss with Circle loss (ablation)SyncTREE (4 layers) · Synthetic datasetslew MAE (ps): 25.672 -> 25.601No clear effect
replace GAT-based baselines with SyncTREE two-pass tree message passingSyncTREE (4 layers) · Synthetic datasetdelay MAE (ps): 9.457 -> 8.745Improved
increase SyncTREE depth from 4 to 32 layersSyncTREE · Synthetic datasetdelay MAE (ps): 8.745 -> 3.424Improved
increase SyncTREE depth from 32 to 64 layersSyncTREE · RISC-V datasetdelay MAE (ps): 0.0106 -> 0.0176Hurt
increase GCN depth from 4 to 64 layersGCN · Synthetic datasetdelay MAE (ps): 12.675 -> 13.782Hurt
transfer pretrained delay representation to slew task with frozen weights and fine-tuned readout (ablation)SyncTREE (16 layers) · Synthetic datasetslew MAE change: 2.304Improved

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