Causal Influence in Federated Edge Inference
This work addresses the need to distinguish causal influence from correlations in cooperative inference systems, which is incremental as it applies a causal framework to a known federated setting.
The paper tackles the problem of evaluating individual agent influence in federated edge inference under uncertainty, deriving causal impact expressions and validating them with simulations and a real-world crowd counting application.
In this paper, we consider a setting where heterogeneous agents with connectivity are performing inference using unlabeled streaming data. Observed data are only partially informative about the target variable of interest. In order to overcome the uncertainty, agents cooperate with each other by exchanging their local inferences with and through a fusion center. To evaluate how each agent influences the overall decision, we adopt a causal framework in order to distinguish the actual influence of agents from mere correlations within the decision-making process. Various scenarios reflecting different agent participation patterns and fusion center policies are investigated. We derive expressions to quantify the causal impact of each agent on the joint decision, which could be beneficial for anticipating and addressing atypical scenarios, such as adversarial attacks or system malfunctions. We validate our theoretical results with numerical simulations and a real-world application of multi-camera crowd counting.