PreRoutGNN for Timing Prediction with Order Preserving Partition: Global Circuit Pre-training, Local Delay Learning and Attentional Cell Modeling (2403.00012v2)
Abstract: Pre-routing timing prediction has been recently studied for evaluating the quality of a candidate cell placement in chip design. It involves directly estimating the timing metrics for both pin-level (slack, slew) and edge-level (net delay, cell delay), without time-consuming routing. However, it often suffers from signal decay and error accumulation due to the long timing paths in large-scale industrial circuits. To address these challenges, we propose a two-stage approach. First, we propose global circuit training to pre-train a graph auto-encoder that learns the global graph embedding from circuit netlist. Second, we use a novel node updating scheme for message passing on GCN, following the topological sorting sequence of the learned graph embedding and circuit graph. This scheme residually models the local time delay between two adjacent pins in the updating sequence, and extracts the lookup table information inside each cell via a new attention mechanism. To handle large-scale circuits efficiently, we introduce an order preserving partition scheme that reduces memory consumption while maintaining the topological dependencies. Experiments on 21 real world circuits achieve a new SOTA R2 of 0.93 for slack prediction, which is significantly surpasses 0.59 by previous SOTA method. Code will be available at: https://github.com/Thinklab-SJTU/EDA-AI.
- OpenROAD: Toward a Self-Driving, Open-Source Digital Layout Implementation Tool Chain. Proc. GOMACTECH.
- Beit: Bert pre-training of image transformers. In ICLR.
- Machine learning-based pre-routing timing prediction with reduced pessimism. In DAC.
- Language models are few-shot learners. NeurIPS.
- Simple and deep graph convolutional networks. In ICML.
- The Policy-gradient Placement and Generative Routing Neural Networks for Chip Design. In NeurIPS.
- On joint learning for solving placement and routing in chip design. In NeurIPS.
- Too Big to Fail? Active Few-Shot Learning Guided Logic Synthesis. arXiv preprint arXiv:2204.02368.
- Convolutional neural networks on graphs with fast localized spectral filtering. In NeurIPS.
- Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
- Hubrouter: Learning global routing via hub generation and pin-hub connection. In NeurIPS.
- Generalizable Cross-Graph Embedding for GNN-based Congestion Prediction. In ICCAD.
- A timing engine inspired graph neural network model for pre-routing slack prediction. In DAC.
- Algorithm AS 136: A k-means clustering algorithm. Journal of the royal statistical society. series c (applied statistics).
- TAU 2015 contest on incremental timing analysis. In ICCAD.
- TAU 2014 contest on removing common path pessimism during timing analysis. In ISPD.
- Strategies for pre-training graph neural networks. In ICLR.
- A software package for partitioning unstructured graphs, partitioning meshes, and computing fill-reducing orderings of sparse matrices. University of Minnesota, Department of Computer Science and Engineering, Army HPC Research Center, Minneapolis, MN.
- DeepSeq: Deep Sequential Circuit Learning. arXiv preprint arXiv:2302.13608.
- Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
- Semi-supervised classification with graph convolutional networks. In ICLR.
- Lasser, D. J. 1961. Topological ordering of a list of randomly-numbered elements of a network. Communications of the ACM.
- Deepgate: Learning neural representations of logic gates. In DAC.
- DREAMPlace 4.0: Timing-driven global placement with momentum-based net weighting. In DATE.
- Dreamplace: Deep learning toolkit-enabled gpu acceleration for modern vlsi placement. In DAC.
- ePlace: Electrostatics-based placement using fast fourier transform and Nesterov’s method. ACM Transactions on Design Automation of Electronic Systems (TODAES).
- Nerf: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM.
- A graph placement methodology for fast chip design. Nature.
- How do vision transformers work? In ICLR.
- Pytorch: An imperative style, high-performance deep learning library. In NeurIPS.
- Think locally, act globally: Highly balanced graph partitioning. In Experimental Algorithms: 12th International Symposium, SEA 2013, Rome, Italy, June 5-7, 2013. Proceedings 12.
- Attention is all you need. In NeurIPS.
- Graph attention networks. In ICLR.
- LHNN: Lattice hypergraph neural network for VLSI congestion prediction. In DAC.
- Deep graph library: A graph-centric, highly-performant package for graph neural networks. arXiv preprint arXiv:1909.01315.
- Anti-oversmoothing in deep vision transformers via the fourier domain analysis: From theory to practice. In ICLR.
- Difformer: Scalable (graph) transformers induced by energy constrained diffusion. In ICLR.
- Nodeformer: A scalable graph structure learning transformer for node classification. In NeurIPS.
- Simplifying and Empowering Transformers for Large-Graph Representations. In NeurIPS.
- RouteNet: Routability prediction for mixed-size designs using convolutional neural network. In ICCAD.
- Net2: A graph attention network method customized for pre-placement net length estimation. In ASP-DAC.
- How powerful are graph neural networks? In ICLR.
- Pre-routing path delay estimation based on transformer and residual framework. In ASP-DAC.
- Versatile Multi-stage Graph Neural Network for Circuit Representation. In NeurIPS.
- EasySO: Exploration-enhanced Reinforcement Learning for Logic Synthesis Sequence Optimization and a Comprehensive RL Environment. In ICCAD.
- GraphGLOW: Universal and Generalizable Structure Learning for Graph Neural Networks. In SIGKDD.
- Heterogeneous Graph Neural Network-based Imitation Learning for Gate Sizing Acceleration. In ICCAD.