1: \begin{abstract}

2: In a recent paper by the same authors, we provided a theoretical

3: foundation for the component-by-component (CBC) construction of lattice

4: algorithms for multivariate $L_2$ approximation in the worst case setting,

5: for functions in a periodic space with general weight parameters. The

6: construction led to an error bound that achieves the best possible rate of

7: convergence for lattice algorithms. Previously available literature

8: covered only weights of a simple form commonly known as \emph{product

9: weights}. In this paper we address the computational aspect of the

10: construction. We develop \emph{fast} CBC construction of lattice

11: algorithms for special forms of weight parameters, including the so-called

12: \emph{POD weights} and \emph{SPOD weights} which arise from PDE

13: applications, making the lattice algorithms truly applicable in practice.

14: With $d$ denoting the dimension and $n$ the number of lattice points, we

15: show that the construction cost is $\calO(d\,n\log(n) + d^2\log(d)\,n)$

16: for POD weights, and $\calO(d\,n\log(n) + d^3\sigma^2\,n)$ for SPOD

17: weights of degree $\sigma\ge 2$. The resulting lattice generating vectors

18: can be used in other lattice-based approximation algorithms, including

19: kernel methods or splines.

20: \\[2mm]

21: \textbf{AMS Subject Classification: }41A10, 41A15, 65D30, 65D32, 65T40.

22: % 41A10   	Approximation by polynomials {For approximation by trigonometric polynomials, see 42A10}

23: % 41A15   	Spline approximation

24: % 41A63   	Multidimensional problems (should also be assigned at least one other classification number in this section)

25: % 42A10   	Trigonometric approximation

26: % 42A15   	Trigonometric interpolation

27: % 65D30   	Numerical integration

28: % 65D32   	Quadrature and cubature formulas

29: % 65T40   	Trigonometric approximation and interpolation

30: \end{abstract}

31: