RESEARCH METHODOLOGY

Research Methodology

Professor Thuan Dinh Do’s research methodology focuses on a hands-on, project-based approach that integrates embedded systems, sensors, and machine learning into practical applications. Students are guided through the complete development pipeline of Tiny Machine Learning systems, beginning with data collection from real-world sensors, followed by data preprocessing and machine learning model training. 

After training the models, students learn techniques for model optimization and compression so that the algorithms can run efficiently on resource-constrained microcontrollers. 

The final stage of the methodology involves deploying and evaluating the models directly on embedded devices, allowing students to observe real-time inference and system performance. This end-to-end process helps learners understand both the theoretical foundations and practical challenges of implementing artificial intelligence at the edge.

TEACHING PHILOSOPHY

Professor Thuan Dinh Do and his lab members selected teaching philosophy which emphasizes experiential learning, accessibility, and innovation. They believes that complex topics such as machine learning and embedded AI can become approachable when students are engaged through hands-on experiments and real-world engineering problems. 

By adapting undergraduate-level curricula from U.S. universities to the high school level, Professor Thuan Dinh Do and his lab members provide students with early exposure to advanced technologies in computer engineering, data science, and artificial intelligence. 

Our approach encourages students to develop critical thinking, creativity, and problem-solving skills, while also building confidence in their ability to design intelligent systems. Ultimately, we aim to inspire the next generation of engineers and researchers by making emerging technologies like Tiny Machine Learning both understandable and exciting for young learners.

FROM COURSES TAUGHT IN THE US

In Professor Thuan Dinh Do’s undergraduate courses in the United States, Professor Do teaches students how to design embedded systems that integrate machine learning, sensors, and microcontrollers. Students learn the entire development process, including data collection, model training, model optimization, and deployment on embedded devices. By adapting this curriculum for high school students, Professor Do makes advanced AI technologies accessible to younger learners who are interested in engineering, computer science, and data science.

Through hands-on activities, students learn to build projects such as gesture recognition systems, sound detection models, and motion-based activity classifiers. These projects demonstrate how machine learning can run directly on small devices like the Arduino Nano 33 BLE Sense, providing real-time intelligence without needing a powerful computer. This practical experience allows students to understand both the theoretical concepts of machine learning and the engineering challenges of implementing AI on embedded hardware.

PREPARING VIETNAMESE STUDENTS FOR THE FUTURE OF ARTIFICIAL INTELLIGENCE

Professor Thuan Dinh Do believes that early exposure to emerging technologies can inspire students to pursue careers in artificial intelligence, embedded systems, and computer engineering. By introducing Vietnamese high school students to TinyML, he hopes to help them build strong technical foundations and prepare for future studies at leading universities around the world, particularly in the United States.

Through these educational initiatives, Professor Thuan Dinh Do is bridging the gap between university-level engineering education and high school learning, giving students the opportunity to explore cutting-edge technologies and develop the skills needed to become the next generation of innovators in artificial intelligence and embedded systems.