# SparseKAN: Compressing Kolmogorov--Arnold Networks Across Basis Functions, Neurons, and Bits

arXiv: https://arxiv.org/abs/2608.00859 · HTML: https://mlsanity.com/paper/2608.00859/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 20 references that are in the map.
- First posted on arXiv 1 Aug 2026.

## Builds on (most cited first, 12 of 12)

- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Learning Efficient Convolutional Networks through Network Slimming](https://mlsanity.com/paper/1708.06519/) (2017)
- [Learning Sparse Neural Networks through L0 Regularization](https://mlsanity.com/paper/1712.01312/) (2017)
- [Chebyshev Polynomial-Based Kolmogorov-Arnold Networks: An Efficient Architecture for Nonlinear Function Approximation](https://mlsanity.com/paper/2405.07200/) (2024)
- [Bayesian Bits: Unifying Quantization and Pruning](https://mlsanity.com/paper/2005.07093/) (2020)
- [Differentiable Joint Pruning and Quantization for Hardware Efficiency](https://mlsanity.com/paper/2007.10463/) (2020)
- [KANEL'E: Kolmogorov-Arnold Networks for Efficient LUT-based Evaluation](https://mlsanity.com/paper/2512.12850/) (2025)
- [Architectural Scaling Surpass Basis Complexity? Efficient KANs with Single-Parameter Design](https://mlsanity.com/paper/2410.14951/) (2024)
- [KANtize: Exploring Low-bit Quantization of Kolmogorov-Arnold Networks for Efficient Inference](https://mlsanity.com/paper/2603.17230/) (2026)
- [FlashKAT: Understanding and Addressing Performance Bottlenecks in the Kolmogorov-Arnold Transformer](https://mlsanity.com/paper/2505.13813/) (2025)
- [KAN-SAs: Efficient Acceleration of Kolmogorov-Arnold Networks on Systolic Arrays](https://mlsanity.com/paper/2512.00055/) (2025)
- [VIKIN: A Reconfigurable Accelerator for KANs and MLPs with Two-Stage Sparsity Support](https://mlsanity.com/paper/2603.01165/) (2026)

## Abstract (arXiv)

Kolmogorov--Arnold Networks (KANs) replace scalar edge weights with learnable univariate functions parameterized by multiple basis coefficients. This introduces a source of redundancy that conventional neural-network compression does not directly expose. We present SparseKAN, a unified approach that compresses KANs along three complementary axes: basis functions, neurons/channels, and numerical precision. SparseKAN equips the base branch, nonlinear basis branch, and individual basis terms with hierarchical learnable gates trained under a differentiable active-cost objective. The learned importance structure is subsequently hardened under explicit basis and width budgets, recovered in full or low precision, and physically compacted into smaller dense tensors rather than retained as sparse masks. Experiments on MNIST, CIFAR-10, and CIFAR-100 across spline, polynomial, RBF, wavelet, and convolutional KAN variants show that the structural axes compose predictably in cost. We also find strong basis-dependent differences in term importance: coefficient-based selection outperforms matched low-order truncation by up to 15.25 accuracy points in the evaluated Gram-polynomial settings. Eight-bit quantization is broadly robust, whereas 4-bit convolutional KANs require quantization-aware adaptation. Physical compaction removes up to 73.0% of parameters without accuracy loss on MNIST and reduces large-batch CUDA latency to as little as 0.51× dense execution. On a ZCU104 FPGA, the resulting sparse low-bit models achieve up to 23.63× lower inference latency, demonstrating that SparseKAN converts functional redundancy into measurable software and hardware efficiency. The SparseKAN implementation is available at https://github.com/OSU-STARLAB/SparseKAN.