Model-based Approach with ACT-R about Benefits of Memory-based Strategy on Anomalous Behaviors

AbstractUsers sometimes face anomalous behaviors of systems, such as machine failures and autonomous agents. Predicting such behaviors of systems is difficult. We investigate the benefits of the memory-based strategy, which focuses on memorization of instances to predict anomalous and regular behaviors of the system, with ACT-R simulations with a cognitive model. In this study, we presumed the parameters defining the encoding processes on anomalous instances and regular instances in the model of the memory-based strategy and performed simulations to verify how these two parameters influence prediction performance. The results of simulations showed that (1) regular instances are not encoded as default values in the memory-based strategy and that (2) such inactivity on regular instances suppresses commission errors of regular instances and does not suppress commission errors of anomalous instances nor omission errors.

Return to previous page