We propose an information-theoretic Markov aggregation framework that is motivated by two objectives: 1) The Markov chain observed through the aggregation mapping should be Markov. 2) The aggregated chain should retain the temporal dependence structure of the original chain. We analyze our parameterized cost function and show that it contains previous cost functions as special cases, which we critically assess. Our simple optimization heuristic for deterministic aggregations characterizes the optimization landscape for different parameter values.
Amjad, R. A., Blöchl, C., & Geiger, B. (2020). A Generalized Framework For Kullback-Leibler Markov Aggregation. IEEE Transactions on Automatic Control, 65(7), 3068-3075. https://doi.org/10.1109/TAC.2019.2945891