January 30, 2023Gleb Pogudin (LIX)
Dynamical systems are frequently used for modeling in the sciences. Building a detailed model often requires taking into account a large number of factors and, as a result, a model may become quite large. High-dimensional models with dozens or hundreds of state variables are not only challenging computationally but it is also hard to use them to derive mechanistic insights. Exact model reduction is a way to address this issue by finding a self-consistent lower dimensional projection of the corresponding dynamical system. I will describe recent algorithms for computing such reductions and demonstrate them on the models from literature.