Kernel Regression with Tensor Trains and Hadamard Overparameterization

arXiv:2607.17390 · stat.ML, cs.LG, eess.SP · Submitted 2026-07-19 · Read on arXiv

Duc Thien Nguyen, Konstantinos Slavakis, Eleftherios Kofidis, Dimitris Pados

stat.ML, cs.LG, eess.SP

Submitted: 2026-07-19

Code: https://github.com/bstabler/TransportationNetworks

License: http://creativecommons.org/licenses/by/4.0/

The gist: Kernel regression with tensor trains and Hadamard overparameterization (KReTTaH) is introduced as a training-data-free, interpretable, and nonparametric framework for multi-way data imputation.

Terminology

Abstract

Kernel regression with tensor trains and Hadamard overparameterization (KReTTaH) is introduced as a training-data-free, interpretable, and nonparametric framework for multi-way data imputation. The imputation problem is reformulated as regression in reproducing kernel Hilbert spaces (RKHS), where the tensor regression coefficients are explicitly constrained to lie on fixed-rank tensor-train (TT) manifolds and structured via Hadamard overparameterization to promote sparsity and high representational efficiency. Rather than relying on costly cross-validation, KReTTaH jointly optimizes the TT coefficient tensors and the kernel covariance matrices within a Riemannian product-manifold framework -- the former on fixed-rank TT manifolds, the latter on the manifold of positive-definite matrices -- thereby enabling automated kernel-hyperparameter selection. Numerical tests on two challenging applications -- imputation of high-dimensional functional magnetic resonance imaging (fMRI) data and recovery of missing edge flows in dynamic graphs -- demonstrate that KReTTaH consistently outperforms state-of-the-art tensor-, Bayesian-, and neural-network-based baselines in terms of modeling accuracy.

Sources

Related papers