Abstract: In this talk, I will discuss how optimization and machine learning can be combined to enhance understanding of human movement. It is well known that humans perform an optimization when planning movements, which makes optimal control theory suitable to simulate human movement. At the same time, machine learning has shown its merit in many different fields, most notable image processing, and has also seen successful application in movement analysis recently. In this talk, I will discuss how we can combine optimal control theory and machine learning to further the field of biomechanical movement analysis. On the one hand, optimal control theory can provide new insights on a full-body scale, for example to predict gait with a newly developed prosthesis, to reconstruct movements from wearable sensor measurement or for data augmentation. On the other hand, data-based models can be used when traditional models are not sufficient, for example to extract hidden information from measurements or to create fast processing algorithms. By combining both methods, we can benefit from the strengths of both methods to develop new insights into human movement that can improve performance, for example for athletes or persons with a gait disability.