Developing new materials lays the foundation for technology innovations. When one material is integrated with another, new properties and functionalities can emerge in the resulting heterostructure. The choice of building blocks, however, is a challenge that is best addressed with a collaborative approach combining computational methods, material synthesis, and a broad range of characterization methods. This research will tackle a long-standing material science challenge: how to create multiferroics materials that combine long-range electric and magnetic orders. Artificial multiferroics consisting of layered magnetic two-dimensional (2D) materials interfaced with ferroelectric 2D or oxides materials will be investigated. Because the interface between the electrically ordered (ferroelectric) layer and magnetically ordered (e.g., ferromagnetic) layer is atomically flat, an enhanced coupling between the two can be used to effectively switch the magnetic order via the electrical control. These new materials can lead to technological innovations, e.g., memory devices that are compact and power-saving. Given the large number of possible choices of 2D materials, machine-learning based data mining will lead the experimental effort in synthesis and characterization of new multiferroic heterostructures in this research.