The Benefit of Decomposing POMDP for Control of Gene Regulatory Networks

Controlling genes and interactions between them is an example real life problem that exhibits partial observability and can be modelled with POMDP framework. In this work, we explore the feasibility of realizing the genes related problem in POMDP framework. Current works addressing partial observability focus on formulating algorithms for the finite horizon gene regulatory network control problem. This motivated us to take the challenge and tackle the control problem from a real infinite horizon partially observable perspective. In other words, the method proposed in this work is a POMDP formulation for the infinite horizon version of the problem. This formulation first decomposes the problem by isolating different unrelated parts of the problem, and then makes use of existing POMDP solvers to solve the obtained sub problems, the final outcome is a control mechanism for the main problem.