Conditioning Gaussian measure on Hilbert space
Houman Owhadi, Clint Scovel

TL;DR
This paper characterizes how Gaussian measures on Hilbert spaces behave under conditioning on subspaces, showing the resulting measures are Gaussian with a specific covariance operator, and develops approximation methods for these operators.
Contribution
It provides new proofs and an approximation framework for the shorted operator in Gaussian measures on Hilbert spaces, extending existing theory.
Findings
Conditional measures are Gaussian with shorted covariance operators.
Develops an approximation scheme for the shorted operator via symmetric oblique projections.
Establishes convergence of approximations in weak and trace norms.
Abstract
For a Gaussian measure on a separable Hilbert space with covariance operator , we show that the family of conditional measures associated with conditioning on a closed subspace are Gaussian with covariance operator the short of the operator to . We provide two proofs. The first uses the theory of Gaussian Hilbert spaces and a characterization of the shorted operator by Andersen and Trapp. The second uses recent developments by Corach, Maestripieri and Stojanoff on the relationship between the shorted operator and -symmetric oblique projections onto . To obtain the assertion when such projections do not exist, we develop an approximation result for the shorted operator by showing, for any positive operator , how to construct a sequence of approximating operators which possess -symmetric oblique projections onto…
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.
Taxonomy
TopicsMarkov Chains and Monte Carlo Methods · Mathematical Approximation and Integration · Numerical methods in inverse problems
