High-entropy alloys, which involve features of both crystalline and amorphous solids, are regarded as a new avenue in search for low thermal conductivity. However, due to their highly disordered nature, understanding and predicting their thermal conductivity from first principles calculations is very challenging. In this master thesis project, which will lead to a fully-funded PhD position, we plan to use machine learning potentials to investigate the thermal conductivity of high-entropy oxides. This will involve evaluating the architecture of available machine learning interatomic potentials, generating the training data using density functional theory calculations, and training the machine learning potentials. Finally, we will use these machine learning potentials in calculating the thermal conductivity and understanding the mechanisms of heat flow.