The classical theory of Gaussian measures on infinite-dimensional spaces was developed by Leonard Gross in the 1960s. His key result, known as the Abstract Wiener Space theorem, explains how to construct well-behaved Gaussian probability measures on certain infinite-dimensional spaces by embedding a smaller, nicer space inside a larger one. The theorem is fundamental to areas like stochastic analysis, quantum field theory, and the study of Brownian motion in abstract settings. The original proof works for Banach spaces, which are infinite-dimensional spaces with a specific kind of distance structure.
This paper extends Gross's theorem to a broader class of spaces called Frechet spaces. Frechet spaces are more general than Banach spaces: instead of having a single notion of distance, they have a whole sequence of ways to measure size and closeness. Examples include spaces of smooth functions, which arise naturally in analysis and physics. Because Frechet spaces lack some of the convenient tools available in Banach spaces, extending the theorem requires reworking the argument carefully.
The authors adapt a proof strategy originally developed by Bogachev for the Banach setting, using a technique called bootstrapping, where you build up a strong conclusion by repeatedly applying a weaker result in a clever sequence of steps. By adapting this approach to the Frechet setting, they not only achieve the generalization but also simplify several parts of the argument compared to earlier approaches. The result makes the Abstract Wiener Space framework available for a wider range of spaces that appear naturally in modern analysis and probability theory.