“…To address above basic and important issue in handling DMOPs, the existing DMOAs can be generally divided into the memory- [45], [59], prediction- [42], [48]- [50], [57], [70], and diversity-based methods [12], [35], [36]. Additionally, some hybrid algorithms have also been proposed [8], [21], [31], [53], [69], and in particular, a novel trend of developing DMOAs has emerged recently, which combines the memory mechanism and the prediction method, where the transfer learning (TL) technique has been adopted to make full use of history knowledge to accelerate convergence in a new environment, see [25], [26], [33], [39], [55], [60] for some successful applications.…”