We will present our paper Fine-Tuning of Neural Network Approximate MPC without Retraining via Bayesian Optimization at the International Conference on Robot Intelligence Technology and Applications (RITA).