ML-Based Automata Simplification for Symbolic Accelerators

2025-07-11

AI Summary

Symbolic accelerators are widely deployed in genomics, NLP, and cybersecurity; however, their NFA-based implementations suffer from high FPGA resource consumption and routing complexity, limiting scalability to large-scale graphs. This paper introduces AutoSlim—the first machine learning–driven graph simplification framework—featuring a novel semantic-preserving edge scoring and pruning mechanism powered by random forests applied to weighted transitions, integrated with structural graph features for low-impact edge identification. AutoSlim is deployable on FPGA overlay architectures such as NAPOLY+ to enable efficient sequence analysis. Experiments across graphs of 1K–64K nodes demonstrate that AutoSlim reduces LUT usage by 40% on average and prunes over 30% of transition edges, overcoming the scalability bottlenecks of existing tools and enabling processing of graphs an order of magnitude larger than state-of-the-art baselines.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Symbolic accelerators are increasingly used for symbolic data processing in domains such as genomics, NLP, and cybersecurity. However, these accelerators face scalability issues due to excessive memory use and routing complexity, especially when targeting a large set. We present AutoSlim, a machine learning-based graph simplification framework designed to reduce the complexity of symbolic accelerators built on Non-deterministic Finite Automata (NFA) deployed on FPGA-based overlays such as NAPOLY+. AutoSlim uses Random Forest classification to prune low-impact transitions based on edge scores and structural features, significantly reducing automata graph density while preserving semantic correctness. Unlike prior tools, AutoSlim targets automated score-aware simplification with weighted transitions, enabling efficient ranking-based sequence analysis. We evaluated data sets (1K to 64K nodes) in NAPOLY+ and conducted performance measurements including latency, throughput, and resource usage. AutoSlim achieves up to 40 percent reduction in FPGA LUTs and over 30 percent pruning in transitions, while scaling to graphs an order of magnitude larger than existing benchmarks. Our results also demonstrate how hardware interconnection (fanout) heavily influences hardware cost and that AutoSlim's pruning mitigates resource blowup.

Problem

• Reduces memory use and routing complexity in symbolic accelerators
• Simplifies Non-deterministic Finite Automata (NFA) graphs using ML
• Improves FPGA resource efficiency and scalability for large datasets

Innovation

• ML-based pruning of low-impact automata transitions
• Random Forest classification for edge scoring
• Score-aware simplification preserving semantic correctness