Research map / Graph learning / Graph self-supervised learning
KCLNet: Electrically Equivalence-Oriented Graph Representation Learning for Analog Circuits
Peng Xu, Yapeng Li, Tinghuan Chen, Tsung-Yi Ho, Bei Yu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use KCL-based positive embeddings instead of augmentation positives (ablation) | KCLNet · ANALOG-DET-242k (subcircuit detection) | mAP: 0.374 -> 0.622 | Improved |
| use KCL-based positive embeddings instead of augmentation positives (ablation) | KCLNet · ANALOG-CLS-428k (circuit classification) | Acc@1: 0.938 -> 0.949 | Improved |
| use KCL-based negatives (node masking) instead of in-batch negatives (ablation) | KCLNet · ANALOG-DET-242k (subcircuit detection) | mAP: 0.602 -> 0.622 | Improved |
| use KCL-based negatives (node masking) instead of in-batch negatives (ablation) | KCLNet · ANALOG-CLS-428k (circuit classification) | F1 score: 0.682 -> 0.853 | Mixed |
| add full KCL loss on top of asynchronous message passing (ablation) | KCLNet · ANALOG-DET-242k (subcircuit detection) | mAP: 0.561 -> 0.622 | Improved |
| add full KCL loss on top of asynchronous message passing (ablation) | KCLNet · ANALOG-CLS-428k (circuit classification) | Acc@1: 0.938 -> 0.949 | Improved |
| replace synchronous GIN encoder with asynchronous message passing (AGNN) (ablation) | AGNN · ANALOG-CLS-428k (circuit classification) | Acc@1: 0.786 -> 0.938 | Improved |
| replace synchronous GIN encoder with asynchronous message passing (AGNN) (ablation) | AGNN · ANALOG-DET-242k (subcircuit detection) | mAP: 0.553 -> 0.561 | Mixed |
| add generic GraphCL pretraining to GIN (ablation) | GIN · ANALOG-DET-242k (subcircuit detection) | mAP: 0.553 -> 0.434 | Hurt |
| add generic GraphCL pretraining to GIN (ablation) | GIN · ANALOG-CLS-428k (circuit classification) | Acc@1: 0.786 -> 0.875 | Improved |
| add generic GraphCL pretraining to GCN (ablation) | GCN · ANALOG-CLS-428k (circuit classification) | Acc@1: 0.561 -> 0.679 | Improved |
| add generic GraphCL pretraining to SAGE (ablation) | SAGE · ANALOG-DET-242k (subcircuit detection) | mAP: 0.355 -> 0.361 | No clear effect |
Closest papers by meaning
- AnalogToBi: Device-Level Analog Circuit Topology Generation via Bipartite Graph and Grammar Guided Decoding (NeurIPS 2026 · similarity 0.88)
- Circuit Representation Learning with Masked Gate Modeling and Verilog-AIG Alignment (ICLR 2025 · similarity 0.88)
- DeepGate4: Efficient and Effective Representation Learning for Circuit Design at Scale (ICLR 2025 · similarity 0.87)
- CircuitFusion: Multimodal Circuit Representation Learning for Agile Chip Design (ICLR 2025 · similarity 0.86)
- FALCON: An ML Framework for Fully Automated Layout-Constrained Analog Circuit Design (NeurIPS 2025 · similarity 0.86)
- AnalogGenie: A Generative Engine for Automatic Discovery of Analog Circuit Topologies (ICLR 2025 · similarity 0.86)
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.
