aiqbench analysis
SNAC-Pack addresses the critical gap between accuracy-focused Neural Architecture Search (NAS) and actual FPGA hardware constraints by integrating a multi-objective surrogate model that estimates resource usage like DSPs and BRAM before synthesis. This open-source framework automates the end-to-end pipeline from global search to quantization-aware training and HLS generation, significantly reducing iteration costs for hardware-aware model design. Its demonstrated utility in high-stakes domains like particle physics and quantum readout suggests strong potential for optimizing edge AI deployments where latency and area are primary bottlenecks.
Surrogate Neural Architecture Codesign Package (SNAC-Pack)
arXiv:2605.16138v2 Announce Type: replace-cross Abstract: Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.
Abstract:Neural architecture search (NAS) is a powerful approach for automating model design, but existing methods often optimize for accuracy alone or rely on proxy metrics such as bit operations (BOPs) that correlate poorly with hardware cost. This gap is particularly large for FPGA deployment, where cost is dominated by a multi-dimensional budget of lookup tables, DSPs, flip-flops, BRAM, and latency. We present the Surrogate Neural Architecture Codesign Package (SNAC-Pack), an open-source AutoML framework for hardware-aware neural architecture codesign and end-to-end FPGA deployment. SNAC-Pack runs a multi-objective global search with Optuna and NSGA-II, loading trials to a shared SQLite store that enables parallel workers across compute nodes. A hardware surrogate model outputs per-trial resource and latency estimates, avoiding the synthesis cost that would otherwise dominate the search loop. A local search stage then applies quantization-aware training (QAT) together with iterative magnitude pruning in a combined compression loop, after which the final model is synthesized to FPGA firmware via the hls4ml Python library. A YAML configuration and an optional agentic frontend let users run the pipeline on new datasets without modifying the framework. We demonstrate SNAC-Pack on jet classification at the Large Hadron Collider and superconducting qubit readout, discovering compact architectures that match or exceed strong baselines on the task metric while reducing FPGA resource utilization and, in the qubit readout case, reducing the design space exploration process from months of manual fine-tuning to hours of automated search.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); High Energy Physics - Experiment (hep-ex)
arXiv:2605.16138 [cs.LG]
(or arXiv:2605.16138v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2605.16138 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 337 words
Related coverage
SNAC-Pack 2.0: Scaled-Out Surrogate Neural Architecture Codesign
arXiv · Jul 31, 2026
NVIDIA GTC Taipei at COMPUTEX: Live Updates on What’s Next in AI
NVIDIA · May 21, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Build an explainable next-best-product recommendation system for banking on AWS
AWS · Jul 24, 2026