Design Rules for Extreme-Edge Scientific Computing on AI Engines

• URL: http://arxiv.org/abs/2604.19106v1
• Date: Tue, 21 Apr 2026 05:33:11 GMT
• Title: Design Rules for Extreme-Edge Scientific Computing on AI Engines
• Abstract summary: Extreme-edge scientific applications use machine learning models to analyze sensor data and make real-time decisions.<n>This work addresses how and when extreme-edge scientific neural networks should be implemented on AI Engines versus programmable logic.
• Score: 5.134879336741922
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Extreme-edge scientific applications use machine learning models to analyze sensor data and make real-time decisions. Their stringent latency and throughput requirements demand small batch sizes and require that model weights remain fully on-chip. Spatial dataflow implementations are common for extreme-edge applications. Spatial dataflow works well for small networks, but it fails to scale to larger models due to inherent resource scaling limitations. AI Engines on modern FPGA SoCs offer a promising alternative with high compute density and additional on-chip memory. However, the architecture, programming model, and performance-scaling behavior of AI Engines differ fundamentally from those of the programmable logic, making direct comparison non-trivial and the benefits of using AI Engines unclear. This work addresses how and when extreme-edge scientific neural networks should be implemented on AI Engines versus programmable logic. We provide systematic architectural characterization and micro-benchmarking and introduce a latency-adjusted resource equivalence (LARE) metric that identifies when AI Engine implementations outperform programmable logic designs. We further propose spatial and API-level dataflow optimizations tailored to low-latency scientific inference. Finally, we demonstrate the successful deployment of end-to-end neural networks on AI Engines that cannot fit on programmable logic when using the hlsml toolchain.

Related papers

• Scaling LLM Speculative Decoding: Non-Autoregressive Forecasting in Large-Batch Scenarios [76.85739138203014]
  We present SpecFormer, a novel architecture that accelerates unidirectional and attention mechanisms.<n>We demonstrate that SpecFormer achieves lower training demands and reduced computational costs.
  arXiv  Detail & Related papers  (2025-11-25T14:20:08Z)
• Model-free front-to-end training of a large high performance laser neural network [0.0]
  We demonstrate a fully autonomous and parallel optical neural network (ONN) using off-the-shelf components.<n>Our ONN is highly efficient and is scalable both in network size and inference bandwidth towards the GHz range.<n>We show that our ONN can achieve high accuracy and convergence efficiency, even under limited hardware resources.
  arXiv  Detail & Related papers  (2025-03-21T08:43:02Z)
• Analysis of Hardware Synthesis Strategies for Machine Learning in Collider Trigger and Data Acquisition [0.0]
  Machine learning can be implemented in detector electronics for intelligent data processing and acquisition. implementation of ML in real-time at colliders requires very low latencies that are unachievable with a software-based approach. An analysis of neural network inference efficiency is presented, focusing on the application of collider trigger algorithms in field programmable gate arrays.
  arXiv  Detail & Related papers  (2024-11-18T15:59:30Z)
• Automatic AI Model Selection for Wireless Systems: Online Learning via Digital Twinning [50.332027356848094]
  AI-based applications are deployed at intelligent controllers to carry out functionalities like scheduling or power control. The mapping between context and AI model parameters is ideally done in a zero-shot fashion. This paper introduces a general methodology for the online optimization of AMS mappings.
  arXiv  Detail & Related papers  (2024-06-22T11:17:50Z)
• Using the Abstract Computer Architecture Description Language to Model AI Hardware Accelerators [77.89070422157178]
  Manufacturers of AI-integrated products face a critical challenge: selecting an accelerator that aligns with their product's performance requirements. The Abstract Computer Architecture Description Language (ACADL) is a concise formalization of computer architecture block diagrams. In this paper, we demonstrate how to use the ACADL to model AI hardware accelerators, use their ACADL description to map DNNs onto them, and explain the timing simulation semantics to gather performance results.
  arXiv  Detail & Related papers  (2024-01-30T19:27:16Z)
• Dynamic Early Exiting Predictive Coding Neural Networks [3.542013483233133]
  With the urge for smaller and more accurate devices, Deep Learning models became too heavy to deploy. We propose a shallow bidirectional network based on predictive coding theory and dynamic early exiting for halting further computations. We achieve comparable accuracy to VGG-16 in image classification on CIFAR-10 with fewer parameters and less computational complexity.
  arXiv  Detail & Related papers  (2023-09-05T08:00:01Z)
• Intelligence Processing Units Accelerate Neuromorphic Learning [52.952192990802345]
  Spiking neural networks (SNNs) have achieved orders of magnitude improvement in terms of energy consumption and latency. We present an IPU-optimized release of our custom SNN Python package, snnTorch.
  arXiv  Detail & Related papers  (2022-11-19T15:44:08Z)
• Complexity-Driven CNN Compression for Resource-constrained Edge AI [1.6114012813668934]
  We propose a novel and computationally efficient pruning pipeline by exploiting the inherent layer-level complexities of CNNs. We define three modes of pruning, namely parameter-aware (PA), FLOPs-aware (FA), and memory-aware (MA), to introduce versatile compression of CNNs.
  arXiv  Detail & Related papers  (2022-08-26T16:01:23Z)
• FPGA-optimized Hardware acceleration for Spiking Neural Networks [69.49429223251178]
  This work presents the development of a hardware accelerator for an SNN, with off-line training, applied to an image recognition task. The design targets a Xilinx Artix-7 FPGA, using in total around the 40% of the available hardware resources. It reduces the classification time by three orders of magnitude, with a small 4.5% impact on the accuracy, if compared to its software, full precision counterpart.
  arXiv  Detail & Related papers  (2022-01-18T13:59:22Z)
• SOLIS -- The MLOps journey from data acquisition to actionable insights [62.997667081978825]
  In this paper we present a unified deployment pipeline and freedom-to-operate approach that supports all requirements while using basic cross-platform tensor framework and script language engines. This approach however does not supply the needed procedures and pipelines for the actual deployment of machine learning capabilities in real production grade systems.
  arXiv  Detail & Related papers  (2021-12-22T14:45:37Z)
• E3NE: An End-to-End Framework for Accelerating Spiking Neural Networks with Emerging Neural Encoding on FPGAs [6.047137174639418]
  End-to-end framework E3NE automates the generation of efficient SNN inference logic for FPGAs. E3NE uses less than 50% of hardware resources and 20% less power, while reducing the latency by an order of magnitude.
  arXiv  Detail & Related papers  (2021-11-19T04:01:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.