Logic Synthesis Optimization with Predictive Self-Supervision via Causal Transformers

2024-09-16

arXiv.org

AI Summary

Existing logic synthesis optimization (LSO) models suffer from poor generalization and overfitting in quality-of-results (QoR) prediction, primarily due to scarcity of publicly available circuit benchmarks and limited representational capacity of graph encoders. To address these challenges, this work proposes a causal Transformer-based joint modeling framework: (i) a cross-modal co-attention mechanism jointly encodes circuit graphs and optimization action sequences, enabling structural-action co-representation; and (ii) a novel predictive self-supervised learning paradigm tailored for data-scarce settings, alleviating graph encoding bottlenecks and mitigating policy overfitting in reinforcement learning. Evaluated on EPFL, OABCD, and a private industrial dataset, the method reduces QoR prediction error by 5.74%, 4.35%, and 17.06%, respectively—outperforming state-of-the-art baselines significantly.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Probabilistic Circuits and Graphical Models
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Contemporary hardware design benefits from the abstraction provided by high-level logic gates, streamlining the implementation of logic circuits. Logic Synthesis Optimization (LSO) operates at one level of abstraction within the Electronic Design Automation (EDA) workflow, targeting improvements in logic circuits with respect to performance metrics such as size and speed in the final layout. Recent trends in the field show a growing interest in leveraging Machine Learning (ML) for EDA, notably through ML-guided logic synthesis utilizing policy-based Reinforcement Learning (RL) methods.Despite these advancements, existing models face challenges such as overfitting and limited generalization, attributed to constrained public circuits and the expressiveness limitations of graph encoders. To address these hurdles, and tackle data scarcity issues, we introduce LSOformer, a novel approach harnessing Autoregressive transformer models and predictive SSL to predict the trajectory of Quality of Results (QoR). LSOformer integrates cross-attention modules to merge insights from circuit graphs and optimization sequences, thereby enhancing prediction accuracy for QoR metrics. Experimental studies validate the effectiveness of LSOformer, showcasing its superior performance over baseline architectures in QoR prediction tasks, where it achieves improvements of 5.74%, 4.35%, and 17.06% on the EPFL, OABCD, and proprietary circuits datasets, respectively, in inductive setup.

Problem

• Improves logic circuit performance using predictive self-supervised learning.
• Addresses overfitting and generalization in machine learning for EDA.
• Enhances Quality of Results prediction with transformer models.

Innovation

• Autoregressive transformer models for QoR prediction
• Predictive self-supervised learning to enhance accuracy
• Cross-attention modules for circuit and sequence insights