mlsanityData to 7 Oct 2026

IronMan: GNN-assisted Design Space Exploration in High-Level Synthesis via Reinforcement Learning

arXiv 2102.08138 ↗First posted 16 Feb 2021cs.AR cs.CL cs.LGcited by 26

Not on the map. Why?

Abstract (arXiv)

Despite the great success of High-Level Synthesis (HLS) tools, we observe several unresolved challenges: 1) the high-level abstraction of programming styles in HLS sometimes conceals optimization opportunities; 2) existing HLS tools do not provide flexible trade-off (Pareto) solutions among different objectives and constraints; 3) the actual quality of the resulting RTL designs is hard to predict. To address these challenges, we propose an end-to-end framework, namelyIronMan. The primary goal is to enable a flexible and automated design space exploration (DSE), to provide either optimal solutions under user-specified constraints, or various trade-offs among different objectives (such as different types of resources, area, and latency). Such DSE either requires tedious manual efforts or is not achievable to attain these goals through existing HLS tools. There are three components in IronMan: 1) GPP, a highly accurate graph-neural-network-based performance and resource predictor; 2) RLMD, a reinforcement-learning-based multi-objective DSE engine that explores the optimal resource allocation strategy, to provide Pareto solutions between different objectives; 3) CT, a code transformer to assist RLMD and GPP, which extracts the data flow graph from original HLS C/C++ and automatically generates synthesizable code with HLS directives. The experimental results show that: 1) GPP achieves high prediction accuracy, reducing prediction errors of HLS tools by 10.9x in resource utilization and 5.7x in timing; 2) RLMD obtains optimal or Pareto solutions that outperform the genetic algorithm and simulated annealing by 12.7% and 12.9%, respectively; 3) IronMan is able to find optimized solutions perfectly matching various DSP constraints, with 2.54x fewer DSPs and up to 6x shorter latency than those of HLS tools while being up to 400x faster than the heuristic algorithms and HLS tools.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 5 of its 22 references that are in the map. First posted on arXiv 16 Feb 2021.

Builds on

5 of 22 references.

References · 5

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Inductive Representation Learning on Large Graphscited by 6,284
  3. 2018Transformations of High-Level Synthesis Codes for High-Performance Computingcited by 17
  4. 2019Pyramid: Machine Learning Framework to Estimate the Optimal Timing and Resource Usage of a High-Level Synthesis Designcited by 17
  5. 2019Machine Learning Based Routing Congestion Prediction in FPGA High-Level Synthesiscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 7; 2023: 1; 2024: 6; 2025: 6; 2026: 3.

26 papers.

Sort:

By area:

  1. 2022Unsupervised Learning for Combinatorial Optimization with Principled Objective Relaxationcited by 32
  2. 2022High-Level Synthesis Performance Prediction using GNNs: Benchmarking, Modeling, and Advancingcited by 25
  3. 2022BaCO: A Fast and Portable Bayesian Compiler Optimization Frameworkcited by 11
  4. 2024Hierarchical Mixture of Experts: Generalizable Learning for High-Level Synthesiscited by 9
  5. 2024HLSFactory: A Framework Empowering High-Level Synthesis Datasets for Machine Learning and Beyondcited by 8
  6. 2025LIFT: LLM-Based Pragma Insertion for HLS via GNN Supervised Fine-Tuningcited by 8
  7. 2022LOSTIN: Logic Optimization via Spatio-Temporal Information with Hybrid Graph Modelscited by 7
  8. 2022Theta-Resonance: A Single-Step Reinforcement Learning Method for Design Space Explorationcited by 4
  9. 2023Verilog-to-PyG -- A Framework for Graph Learning and Augmentation on RTL Designscited by 4
  10. 2025MetaML-Pro: Cross-Stage Design Flow Automation for Efficient Deep Learning Accelerationcited by 4
  11. 2022Graph Neural Networks: A Powerful and Versatile Tool for Advancing Design, Reliability, and Security of ICscited by 3
  12. 2024Learning to Compare Hardware Designs for High-Level Synthesiscited by 3
14 more
  1. 2024BoolGebra: Attributed Graph-learning for Boolean Algebraic Manipulationcited by 2
  2. 2024Subequivariant Reinforcement Learning Framework for Coordinated Motion Controlcited by 2
  3. 2021Program-to-Circuit: Exploiting GNNs for Program Representation and Circuit Translationcited by 1
  4. 2022A Survey and Perspective on Artificial Intelligence for Security-Aware Electronic Design Automationcited by 1
  5. 2024A Benchmark on Directed Graph Representation Learning in Hardware Designscited by 1
  6. 2025ForgeHLS: A Large-Scale, Open-Source Dataset for High-Level Synthesiscited by 1
  7. 2026MPM-LLM4DSE: Reaching the Pareto Frontier in HLS with Multimodal Learning and LLM-Driven Explorationcited by 1
  8. 2026Agent Factories for High Level Synthesis: How Far Can General-Purpose Coding Agents Go in Hardware Optimization?cited by 1
  9. 2021Pyxis: An Open-Source Performance Dataset of Sparse Acceleratorscited by 0
  10. 2021Enabling Automated FPGA Accelerator Optimization Using Graph Neural Networkscited by 0
  11. 2025TransPlace: Transferable Circuit Global Placement via Graph Neural Networkcited by 0
  12. 2025Ralts: Robust Aggregation for Enhancing Graph Neural Network Resilience on Bit-flip Errorscited by 0
  13. 2025DAPO: Design Structure-Aware Pass Ordering in High-Level Synthesis with Graph Contrastive and Reinforcement Learningcited by 0
  14. 2026SyntheticHLS: Building Diverse Synthetic High-Level Synthesis Datasets using LLMscited by 0