Proving in numerical analysis usually requires both ingenuity and
technical proficiency (e.g., in matrix calculus). The goal of the thesis will
be to free researchers in numerical analysis from some of the more technical
aspects of proving, allowing them to concentrate on the creative aspects of
their work. The thesis work will start with an analysis of several important
articles in the field of numerical analysis. Based on this, we will identify
relevant techniques from automated reasoning and, based on that, build a tool
for supporting researchers in their everyday proof work.
