mlsanityData to 7 Oct 2026

Ultrafast On-Chip Online Learning via Spline Locality in Kolmogorov-Arnold Networks

arXiv 2602.02056 ↗First posted 2 Feb 2026cs.AR cs.LG cs.SYcited by 0

Not on the map. Why?

Abstract (arXiv)

Ultrafast online learning is essential for high-frequency systems, such as controls for quantum computing and nuclear fusion, where adaptation must occur on sub-microsecond timescales. Meeting these requirements demands low-latency, fixed-precision computation under strict memory constraints, a regime in which conventional Multi-Layer Perceptrons (MLPs) are both inefficient and numerically unstable. We identify key properties of Kolmogorov-Arnold Networks (KANs) that align with these constraints. Specifically, we show that: (i) KAN updates exploiting B-spline locality are sparse, enabling superior on-chip resource scaling, and (ii) KANs are inherently robust to fixed-point quantization. By implementing fixed-point online training on Field-Programmable Gate Arrays (FPGAs), a representative platform for on-chip computation, we demonstrate that KAN-based online learners are significantly more efficient and expressive than MLPs across a range of low-latency and resource-constrained tasks. To our knowledge, this work is the first to demonstrate model-free online learning at sub-microsecond latencies.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 24 of its 39 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

24 of 39 references.

References · 24

  1. 2016Layer Normalizationcited by 5,579
  2. 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
  3. 2020The Hardware Lotterycited by 160
  4. 2015All you need is a good initcited by 124
  5. 2018Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RLcited by 80
  6. 2021hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devicescited by 61
  7. 2024Kolmogorov-Arnold Networks: A Critical Assessment of Claims, Performance, and Practical Viabilitycited by 23
  8. 2024Exploring the Limitations of Kolmogorov-Arnold Networks in Classification: Insights to Software Training and Hardware Implementationcited by 22
  9. 2025KANEL'E: Kolmogorov-Arnold Networks for Efficient LUT-based Evaluationcited by 13
  10. 2024Hardware Acceleration of Kolmogorov-Arnold Network (KAN) for Lightweight Edge Inferencecited by 11
  11. 2021On-FPGA Training with Ultra Memory Reduction: A Low-Precision Tensor Methodcited by 9
  12. 2022EF-Train: Enable Efficient On-device CNN Training on FPGA Through Data Reshaping for Online Adaptation or Personalizationcited by 8
12 more
  1. 2025End-to-end workflow for machine learning-based qubit readout with QICK and hls4mlcited by 8
  2. 2023Low latency optical-based mode tracking with machine learning deployed on FPGAs on a tokamakcited by 7
  3. 2024A survey on FPGA-based accelerator for ML modelscited by 7
  4. 2019Automatic Compiler Based FPGA Accelerator for CNN Trainingcited by 6
  5. 2017Learning Sparse Representations in Reinforcement Learning with Sparse Codingcited by 5
  6. 2018A Highly Parallel FPGA Implementation of Sparse Neural Network Trainingcited by 4
  7. 2022FPGA Implementation of Multi-Layer Machine Learning Equalizer with On-Chip Trainingcited by 2
  8. 2024FPGA-based Distributed Union-Find Decoder for Surface Codescited by 2
  9. 2023Efficient Implementation of a Multi-Layer Gradient-Free Online-Trainable Spiking Neural Network on FPGAcited by 1
  10. 2024Demonstrating the Potential of Adaptive LMS Filtering on FPGA-Based Qubit Control Platforms for Improved Qubit Readout in 2D and 3D Quantum Processing Unitscited by 1
  11. 2025Function Approximation Using Analog Building Blocks in Flexible Electronicscited by 1
  12. 2025Online Learning Extreme Learning Machine with Low-Complexity Predictive Plasticity Rule and FPGA Implementationcited by 1

Built on by

Not cited yet.