This paper describes the artificial epigenetic network (AEN), a recurrent connectionist architecture that is able to dynamically modify its topology in order to automatically decompose and solve dynamical problems. The approach is motivated by the behaviour of gene regulatory networks, particularly the epigenetic process of chromatin remodelling that leads to topological change and which underlies the differentiation of cells within complex biological organisms. We expected this approach to be useful in situations where there is a need to switch between different dynamical behaviours, and do so in a sensitive and robust manner in the absence of a priori information about problem structure. This hypothesis was tested using a series of dynamical control tasks, each requiring solutions which could express different dynamical behaviours at different stages within the task. In each case, the addition of topological self-modification was shown to improve the performance and robustness of controllers. We believe this is due to the ability of topological changes to stabilise attractors, promoting stability within a dynamical regime whilst allowing rapid switching between different regimes. Post hoc analysis of the controllers also demonstrated how the partitioning of the networks could provide new insights into problem structure.
open access | doi:10.1109/TNNLS.2015.2497142
@article(Turner2017-AENs, author = "Alexander P. Turner and Leo S. D. Caves and Susan Stepney and Andy M. Tyrrell and Michael A. Lones", title = "Artificial Epigenetic Networks: Automatic Decomposition of Dynamical Control Tasks using Topological Self-Modification", journal = "IEEE Transactions on Neural Networks and Learning Systems", volume = 28, number = 1, pages = "218-230", doi = "10.1109/TNNLS.2015.2497142", year = 2017 )