Memory Effect, Rejuvenation and Chaos Effect in the Multi-layer Random Energy Model

We introduce magnetization to the Multi-layer Random Energy Model which has a hierarchical structure, and perform Monte Carlo simulation to observe the behavior of ac-susceptibility. We find that this model is able to reproduce three prominent features of spin glasses, i.e., memory effect, rejuvenation and chaos effect, which were found recently by various experiments on aging phenomena with temperature variations.