arXiv:2602.22422v2 Announce Type: replace-cross Abstract: Smooth-basis models such as Chebyshev polynomial regressors and radial basis function (RBF) networks are well established in numerical analysis. Their continuously differentiable prediction surfaces suit surrogate optimisation, sensitivity analysis, and other settings where the response varies gradually with inputs. Despite these properties, smooth models seldom appear in tabular regression, where tree ensembles dominate. We ask whether they can compete, benchmarking models across 55 regression datasets organised by application domain.
The continuous drive for more efficient and accurate AI models, especially for tabular data where tree ensembles dominate, is pushing researchers to revisit established numerical analysis techniques.
Improving the performance of smooth-basis models for tabular data could offer benefits like better interpretability, continuous differentiability for specific applications, and potentially more robust predictions in certain domains.
This research suggests a potential shift towards incorporating classical numerical methods into modern machine learning, challenging the dominance of tree ensembles in tabular regression applications.
- · AI researchers
- · Industries requiring surrogate optimization
- · Fields needing differentiable predictive models
- · Exclusive reliance on tree ensemble model developers
This research could lead to renewed interest and investment in classical numerical methods for machine learning tasks.
It might encourage the development of hybrid models combining the strengths of both smooth-basis functions and modern machine learning techniques.
These advancements could make AI models more accessible and interpretable for applications where 'black box' models are less desirable.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI