Synergy in Multi-Level Reasoning
Alan N. Steinberg, Christopher L. Bowman · 2022
This paper explores inference patterns within and across the data fusion levels. Methods and examples are presented for synergistic reasoning across diverse entities and situations in a dynamic, complex, and uncertain world. The series of JDL fusion levels often provides a useful and natural sequence for inferencing by component-wise composition: reasoning from measurements to features to states of individuals to relationships, situations, and scenarios. However, there are instances in which other inference sequences are preferable. These include cases where contextual information, provided by the encompassing situation and scenario, can provide expectations or to resolve ambiguities. Inference patterns are identified whereby fusion nodes output data that are at the same, or higher, or lower level than their inputs. A counter-piracy application is examined to illustrate methods for synergistic reasoning within and across fusion levels and the interplay between higher-level fusion and response management in dynamic threat situations.