The unstructured data such as visual information, natural language, and human behaviors opens up a wide array of opportunities in the field of artificial intelligence (AI). The memory-centric neuromorphic computing (MNC) has been proposed for the efficient processing of unstructured data, bypassing the von Neumann bottleneck of current computing architecture. The development of MNC would provide massively parallel processing of unstructured data, realizing the cognitive AI in edge and wearable systems. In this review, recent advances in memory-centric neuromorphic devices are discussed in terms of emerging nonvolatile memories, volatile switches, synaptic plasticity, neuronal models, and memristive neural network.

Controlling the properties of piezoelectric thin films is a key aspect for designing highly efficient flexible electromechanical devices. In this study, the crystallographic phenomena of PbZr_1–xTi_xO₃ (PZT) thin films caused by distinguished interfacial effects are deeply investigated by overlooking views, including not only an experimental demonstration but also ab initio modeling. The polymorphic phase balance and crystallinity, as well as the crystal orientation of PZT thin films at the morphotropic phase boundary (MPB), can be stably modulated using interfacial crystal structures. Here, interactions with MgO stabilize the PZT crystallographic system well and induce the texturing influences, while the PZT film remains quasi-stable on a conventional Al₂O₃ wafer. On the basis of this fundamental understanding, a high-output flexible energy harvester is developed using the controlled-PZT system, which shows significantly higher performance than the unmodified PZT generator. The voltage, current, and power densities are improved by 556%, 503%, and 822%, respectively, in comparison with the previous flexional single-crystalline piezoelectric device. Finally, the improved flexible generator is applied to harvest tiny vibrational energy from a real traffic system, and it is used to operate a commercial electronic unit. These results clearly indicate that atomic-scale designs can produce significant impacts on macroscopic applications.