Feedback loops are circular chains of cause and effect that connect stocks and flows in a system. They can either amplify or dampen the changes in the system, depending on their polarity.

Self-adaptive software systems can adjust their behaviors at runtime to respond to the context changes. To operationalize the adaptive mechanism, feedback loops have been advocated in many works.

Feedback has long been a mysterious effect that has either fixed or ruined circuits. Phase- and loop-gain analysis have served to maintain this view of feedback by being difficult to handle and to ...