A fast and oblivious matrix compression algorithm for Volterra integral operators

TL;DR

This paper introduces a fast, memory-efficient algorithm for computing discretized Volterra integral operators, leveraging hierarchical matrix techniques to improve speed and reduce memory usage in dynamical systems with memory effects.

Contribution

It develops an $ ext{H}^2$-matrix based method for Volterra operators, enabling $O(N)$ complexity and $O( ext{log} N)$ memory, surpassing previous approaches.

Findings

Achieves $O(N)$ computational complexity.

Requires only $O( ext{log} N)$ active memory.

Applicable to a broad class of kernels.

Abstract

The numerical solution of dynamical systems with memory requires the efficient evaluation of Volterra integral operators in an evolutionary manner. After appropriate discretisation, the basic problem can be represented as a matrix-vector product with a lower diagonal but densely populated matrix. For typical applications, like fractional diffusion or large scale dynamical systems with delay, the memory cost for storing the matrix approximations and complete history of the data then would become prohibitive for an accurate numerical approximation. For Volterra-integral operators of convolution type, the \emph{fast and oblivious convolution quadrature} method of Sch\"adle, Lopez-Fernandez, and Lubich allows to compute the discretized valuation with $N$ time steps in $O(N \log N)$ complexity and only requiring $O(\log N)$ active memory to store a compressed version of the complete history of the data. We will show that this algorithm can be interpreted as an $\mathcal{H}$-matrix approximation of the underlying integral operator and, consequently, a further improvement can be achieved, in principle, by resorting to $\mathcal{H}^2$-matrix compression techniques. We formulate a variant of the $\mathcal{H}^2$-matrix vector product for discretized Volterra integral operators that can be performed in an evolutionary and oblivious manner and requires only $O(N)$ operations and $O(\log N)$ active memory. In addition to the acceleration, more general asymptotically smooth kernels can be treated and the algorithm does not require a-priori knowledge of the number of time steps. The efficiency of the proposed method is demonstrated by application to some typical test problems.