Develop efficient and robust empirical estimation of e-machines for large systems

Develop efficient and robust algorithms to estimate e-machines (causal states and transition structure) from empirical data of large complex systems, enabling practical application of the proposed computational closure framework at scale.

Background

The authors highlight that applying the framework to empirical data at scale is challenging primarily due to the difficulty of estimating large e-machines. They note promising directions but do not provide a complete solution.

Creating scalable estimation procedures would bridge theory and practice, facilitating discovery of emergent computational structures in real-world complex systems.