Vietnamese (Native), English (Fluent) Develop novel and efficient machine learning models to process the large‑scale dataset PCQM4Mv2, which contains more than 3 million molecules Optimize deep learning models to reduce the computation cost by 10% ‑ 20% Improve the architecture and setting of existing machine learning models for graphs to boost the accuracy by 5% ‑ 10% Implemented the Yolo V1, an advanced model for object detection, to detect various objects such as cars, motorbikes, and traffic lights on a provided real‑world image dataset Improved the implemented model with recommended tricks to boost the accuracy to 96% Analyzed the model performance and analyzed types of objects that can create difficulties for the model to detect Implemented the Siamese Net, a well‑known Computer Vision model for Object Detection, to track real‑world objects in a provided video dataset Optimize the implemented model to make it more efficient and reduce the inference time by 15 Devel...