On invariant graph subspaces
Konstantin A. Makarov, Stephan Schmitz, Albrecht Seelmann

TL;DR
This paper investigates the decomposition of unbounded 2x2 operator matrices into invariant graph subspaces, providing conditions for such decompositions and introducing a new block diagonalization method.
Contribution
It establishes necessary and sufficient conditions for decomposing operator matrices via invariant graph subspaces and proposes a novel block diagonalization procedure.
Findings
Decomposition occurs if the domain is invariant for angular operators.
A new block diagonalization method is introduced.
Block triangular forms are obtained with a single invariant graph subspace.
Abstract
In this paper we discuss the problem of decomposition for unbounded operator matrices by a pair of complementary invariant graph subspaces. Under mild additional assumptions, we show that such a pair of subspaces decomposes the operator matrix if and only if its domain is invariant for the angular operators associated with the graphs. As a byproduct of our considerations, we suggest a new block diagonalization procedure that resolves related domain issues. In the case when only a single invariant graph subspace is available, we obtain block triangular representations for the operator matrices.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
