Picture of person typing on laptop with programming code visible on the laptop screen

World class computing and information science research at Strathclyde...

The Strathprints institutional repository is a digital archive of University of Strathclyde's Open Access research outputs. Strathprints provides access to thousands of Open Access research papers by University of Strathclyde researchers, including by researchers from the Department of Computer & Information Sciences involved in mathematically structured programming, similarity and metric search, computer security, software systems, combinatronics and digital health.

The Department also includes the iSchool Research Group, which performs leading research into socio-technical phenomena and topics such as information retrieval and information seeking behaviour.


Nonlinear approximation from differentiable piecewise polynomials

Davydov, O. and Petrushev, P. (2003) Nonlinear approximation from differentiable piecewise polynomials. SIAM Journal on Mathematical Analysis, 35 (3). pp. 708-758. ISSN 0036-1410

Full text not available in this repository. Request a copy from the Strathclyde author


We study nonlinear $n$-term approximation in $L_p({mathbb R}^2)$ ($0<pleinfty$) from hierarchical sequences of stable local bases consisting of differentiable (i.e., $C^r$ with $rge1$) piecewise polynomials (splines). We construct such sequences of bases over multilevel nested triangulations of ${mathbb R}^2$, which allow arbitrarily sharp angles. To quantize nonlinear n-term spline approximation, we introduce and explore a collection of smoothness spaces (B-spaces). We utilize the B-spaces to prove companion Jackson and Bernstein estimates and then characterize the rates of approximation by interpolation. Even when applied on uniform triangulations with well-known families of basis functions such as box splines, these results give a more complete characterization of the approximation rates than the existing ones involving Besov spaces. Our results can easily be extended to properly defined multilevel triangulations in ${mathbb R}^d$, d>2.