In evolutionary computation, distinct clusters that address different subproblems evolve independently of each other, which makes it difficult to exchange genetic information between them. However, a vaguely defined task within one system may be expressed more clearly within another. Effective interaction methods enable subsystems to collaborate more effectively in solving global tasks. By analysing how ambiguous intentions regarding electricity consumption influence actual behaviour in real-world scenarios, we discovered that transaction and negotiation patterns within electricity markets can effectively support this process. By introducing time and third parties, the study presents a semiautomatic, interpretable reasoning community logic system that enables machines to express transaction negotiation patterns. Through formalised operations, it facilitates the conversion of intentions, uncovering hidden relationships within global structures through this liberated form of expression. This paper examines its impact on computational and search paradigms through case studies, enabling collaborative approaches and granularity control via dynamic anchor points, and explores automated peer-to-peer transactions and electricity monetisation within highly abstracted power trading processes.
Chen et al. (Mon,) studied this question.