RecKAN: Kolmogorov-Arnold Networks with a Learnable Recursive Polynomial Basis

• URL: http://arxiv.org/abs/2609.01729v1
• Date: Tue, 01 Sep 2026 18:01:32 GMT
• Title: RecKAN: Kolmogorov-Arnold Networks with a Learnable Recursive Polynomial Basis
• Abstract summary: We introduce RecKAN, which defines the basis itself by a second order recurrence, $KAN_Rn+1(x) = (axbx+c) + (dxbx+c) + (dxn-1(x)$, whose five coefficients are learned.<n>We show that its degree grows linearly in $n$ exactly on the sub-family containing all of them, giving a concrete sense that learned basis can move beyond any fixed classical choice.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Kolmogorov--Arnold Networks (KANs) replace the fixed scalar weights of a standard network with learnable univariate functions on each edge, but existing variants still fix the \emph{basis} that those functions are built from: B-splines, Chebyshev polynomials, wavelets, or Jacobi polynomials, and learn only the combination weights over it. We introduce RecKAN, which instead defines the basis itself by a second order polynomial recurrence, $R_{n+1}(x) = (ax^2+bx+c)R_n(x) + (dx+e)R_{n-1}(x)$, whose five coefficients are learned jointly with the network. We show this recurrence recovers several classical polynomial families including both kinds of Chebyshev polynomials, Fibonacci, Pell, and Jacobsthal polynomials as special cases, and prove that its degree grows linearly in $n$ exactly on the sub-family containing all of them, giving a concrete sense in which the learned basis can move beyond any fixed classical choice. Across multiple benchmark datasets spanning image, text, biomedical time series classification, and time series forecasting, RecKAN outperforms three parameter-matched KAN baselines (Chebyshev, Jacobi, and spline based) on all classification tasks and achieves the lowest MSE on the ETTh1 forecasting benchmark. Additionally, when used as a classifier head with a convolutional backbone, RecKAN achieves higher accuracy than standard MLP heads on Fashion MNIST, CIFAR-10, and SVHN. On a synthetic function fitting benchmark it tracks a sharply oscillatory target that a parameter comparable MLP under fits. We further show that the learned recurrence coefficients are interpretable: on the task requiring the most local structure, training moves the basis away from the linear degree growth regime that contains every classical family we identify, consistent with our theoretical analysis of what that structural shift enables.

Related papers

• Primitive Recursion without Composition: Dynamical Characterizations, from Neural Networks to Polynomial ODEs [0.0]
  We study computation-real-valued states evolved by real-valued dynamics.<n>All three operate over continuum-real-valued states evolved by real-valued dynamics.
  arXiv  Detail & Related papers  (2026-04-27T11:48:49Z)
• Beyond Softmax: A Natural Parameterization for Categorical Random Variables [61.709831225296305]
  We introduce the $textitcatnat$ function, a function composed of a sequence of hierarchical binary splits.<n>A rich set of experiments show that the proposed function improves the learning efficiency and yields models characterized by consistently higher test performance.
  arXiv  Detail & Related papers  (2025-09-29T12:55:50Z)
• Ehrenfeucht-Haussler Rank and Chain of Thought [51.33559894954108]
  We present a novel characterization of rank, grounded in the well-known Transformer architecture.<n>We show that the rank of a function $f$ corresponds to the minimum number of emphChain of Thought steps required by a single-layer Transformer.<n>We also introduce the notion of the multi-head rank that captures multi-head single-layer transformers, and perform the analysis of PAC-learnability of the classes of functions with bounded multi-head rank.
  arXiv  Detail & Related papers  (2025-01-22T16:30:58Z)
• Generalization Bounds and Model Complexity for Kolmogorov-Arnold Networks [1.5850926890180461]
  Kolmogorov-Arnold Network (KAN) is a network structure recently proposed by Liu et al.<n>Work provides a rigorous theoretical analysis of KAN by establishing generalization bounds for KAN equipped with activation functions.
  arXiv  Detail & Related papers  (2024-10-10T15:23:21Z)
• Incorporating Arbitrary Matrix Group Equivariance into KANs [69.30866522377694]
  Kolmogorov-Arnold Networks (KANs) have seen great success in scientific domains.<n>We propose Equivariant Kolmogorov-Arnold Networks (EKAN) to broaden their applicability to more fields.
  arXiv  Detail & Related papers  (2024-10-01T06:34:58Z)
• Exploring the Potential of Polynomial Basis Functions in Kolmogorov-Arnold Networks: A Comparative Study of Different Groups of Polynomials [0.0]
  This paper presents a survey of 18 distincts and their potential applications in Gottliebmogorov Network (KAN) models. The study aims to investigate the suitability of theseDistincts as basis functions in KAN models for complex tasks like handwritten digit classification. The performance metrics of the KAN models, including overall accuracy, Kappa, and F1 score, are evaluated and compared.
  arXiv  Detail & Related papers  (2024-05-30T20:40:16Z)
• GINN-LP: A Growing Interpretable Neural Network for Discovering Multivariate Laurent Polynomial Equations [1.1142444517901018]
  We propose GINN-LP, an interpretable neural network, to discover the form of a Laurent Polynomial equation. To the best of our knowledge, this is the first neural network that can discover arbitrary terms without any prior information on the order. We show that GINN-LP outperforms the state-of-theart symbolic regression methods on datasets.
  arXiv  Detail & Related papers  (2023-12-18T03:44:29Z)
• A Unified Algebraic Perspective on Lipschitz Neural Networks [88.14073994459586]
  This paper introduces a novel perspective unifying various types of 1-Lipschitz neural networks. We show that many existing techniques can be derived and generalized via finding analytical solutions of a common semidefinite programming (SDP) condition. Our approach, called SDP-based Lipschitz Layers (SLL), allows us to design non-trivial yet efficient generalization of convex potential layers.
  arXiv  Detail & Related papers  (2023-03-06T14:31:09Z)
• Equivariance with Learned Canonicalization Functions [77.32483958400282]
  We show that learning a small neural network to perform canonicalization is better than using predefineds. Our experiments show that learning the canonicalization function is competitive with existing techniques for learning equivariant functions across many tasks.
  arXiv  Detail & Related papers  (2022-11-11T21:58:15Z)
• Regular Polytope Networks [29.44144177954405]
  We argue that a transformation can be fixed with no loss of accuracy and with a reduction in memory usage. It can also be used to learn stationary and maximally separated embeddings. We show that the stationarity of the embedding and its maximal separated representation can be theoretically justified.
  arXiv  Detail & Related papers  (2021-03-29T14:11:32Z)
• Bilinear Classes: A Structural Framework for Provable Generalization in RL [119.42509700822484]
  Bilinear Classes is a new structural framework which permits generalization in reinforcement learning. The framework incorporates nearly all existing models in which a sample complexity is achievable. Our main result provides an RL algorithm which has sample complexity for Bilinear Classes.
  arXiv  Detail & Related papers  (2021-03-19T16:34:20Z)

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.