Lutz Kämmerer

Lutz Kämmerer, Daniel Potts, Toni Volkmer

The reconstruction of high-dimensional sparse signals is a challenging task in a wide range of applications. In order to deal with high-dimensional problems, efficient sparse fast Fourier transform algorithms are essential tools. The second and third authors have recently proposed a dimension-incremental approach, which only scales almost linear in the number of required sampling values and almost quadratic in the arithmetic complexity with respect to the spatial dimension $d$. Using reconstructing rank-1 lattices as sampling scheme, the method showed reliable reconstruction results in numerical tests but suffers from relatively large numbers of samples and arithmetic operations. Combining the preferable properties of reconstructing rank-1 lattices with small sample and arithmetic complexities, the first author developed the concept of multiple rank-1 lattices. In this paper, both concepts - dimension-incremental reconstruction and multiple rank-1 lattices - are coupled, which yields a distinctly improved high-dimensional sparse fast Fourier transform. Moreover, the resulting algorithm is analyzed in detail with respect to success probability, number of required samples, and arithmetic complexity. In comparison to single rank-1 lattices, the utilization of multiple rank-1 lattices results in a reduction in the complexities by an almost linear factor with respect to the sparsity. Various numerical tests confirm the theoretical results, the high performance, and the reliability of the proposed method.

Lutz Kämmerer

The paper discusses the construction of high dimensional spatial discretizations for arbitrary multivariate trigonometric polynomials, where the frequency support of the trigonometric polynomial is known. We suggest a construction based on the union of several rank-1 lattices as sampling scheme and call such schemes multiple rank-1 lattices. This approach automatically makes available a fast discrete Fourier transform (FFT) on the data. The key objective of the construction of spatial discretizations is the unique reconstruction of the trigonometric polynomial using the sampling values at the sampling nodes. We develop construction methods for multiple rank-1 lattices that allow for this unique reconstruction and for estimates of the number $M$ of distinct sampling nodes within the resulting spatial discretizations. Assuming that the multivariate trigonometric polynomial under consideration is a linear combination of $T$ trigonometric monomials, the oversampling factor $M/T$ is independent of the spatial dimension and, roughly speaking, with high probability only logarithmic in $T$, which is much better than the oversampling factor that is expected when using one single rank-1 lattice. The newly developed approaches for the construction of spatial discretizations are probabilistic methods. The arithmetic complexity of these algorithms depend only linearly on the spatial dimension and, with high probability, only linearly on $T$ up to some logarithmic factors. Furthermore, we analyze the computational complexities of the resulting FFT algorithms in detail and obtain upper bounds in $\mathcal{O}\left(M\log M\right)$, where the constants depend only linearly on the spatial dimension. With high probability, we construct spatial discretizations where $M/T\le C\log T$ holds, which implies that the complexity of the corresponding FFT converts to $\mathcal{O}\left(T\log^2 T\right)$.

Lutz Kämmerer, Toni Volkmer

In this work, we consider the approximate reconstruction of high-dimensional periodic functions based on sampling values. As sampling schemes, we utilize so-called reconstructing multiple rank-1 lattices, which combine several preferable properties such as easy constructability, the existence of high-dimensional fast Fourier transform algorithms, high reliability, and low oversampling factors. Especially, we show error estimates for functions from Sobolev Hilbert spaces of generalized mixed smoothness. For instance, when measuring the sampling error in the $L_2$-norm, we show sampling error estimates where the exponent of the main part reaches those of the optimal sampling rate except for an offset of $1/2+\varepsilon$, i.e., the exponent is almost a factor of two better up to the mentioned offset compared to single rank-1 lattice sampling. Various numerical tests in medium and high dimensions demonstrate the high performance and confirm the obtained theoretical results of multiple rank-1 lattice sampling.

Glenn Byrenheid, Lutz Kämmerer, Tino Ullrich et al.

We consider the approximate recovery of multivariate periodic functions from a discrete set of function values taken on a rank-$s$ integration lattice. The main result is the fact that any (non-)linear reconstruction algorithm taking function values on a rank-$s$ lattice of size $M$ has a dimension-independent lower bound of $2^{-(α+1)/2} M^{-α/2}$ when considering the optimal worst-case error with respect to function spaces of (hybrid) mixed smoothness $α>0$ on the $d$-torus. We complement this lower bound with upper bounds that coincide up to logarithmic terms. These upper bounds are obtained by a detailed analysis of a rank-1 lattice sampling strategy, where the rank-1 lattices are constructed by a component-by-component (CBC) method. This improves on earlier results obtained in [25] and [27]. The lattice (group) structure allows for an efficient approximation of the underlying function from its sampled values using a single one-dimensional fast Fourier transform. This is one reason why these algorithms keep attracting significant interest. We compare our results to recent (almost) optimal methods based upon samples on sparse grids.