TINYTENSOR 2025

TINY MACHINE LEARNING BOOTCAMP 2025 (TINY TENSOR)

Date: 28, 29/6/2025

Venue: Hanoi, Vietnam

Address: 545 Vu Tong Phan Street, Thanh Xuan Dist., Ha Noi, Vietnam

In recent years, machine learning has become one of the most important technologies shaping modern society. From voice assistants and smart homes to autonomous vehicles and healthcare applications, machine learning systems are everywhere. To help high school students explore this exciting field early, a hands-on Tiny Machine Learning Bootcamp (TINY TENSOR) can provide an engaging introduction to artificial intelligence through practical experiments and real-world projects.

The bootcamp uses the Arduino Nano 33 BLE Sense, a powerful microcontroller board designed for edge AI and Tiny Machine Learning (TinyML) applications. This board is fully compatible with the Arduino IDE, making it easy for beginners to write, upload, and test programs. One of its key advantages is that it includes several on-board sensors, such as motion sensors, a microphone, and environmental sensors. These sensors allow students to collect real-world data and use it to train machine learning models.

At the heart of the Arduino Nano 33 BLE Sense is the ARM Cortex-M4 microcontroller, which is powerful enough to run machine learning models directly on the device. This capability enables students to learn about edge computing, where AI models run locally on small devices instead of relying on cloud servers. During the bootcamp, students will learn how models such as wake word detection, keyword recognition, gesture recognition, and anomaly detection can operate directly on embedded systems.

Throughout the program, students will participate in several hands-on projects. For example, they may train a model that recognizes different hand movements using the board’s motion sensors, or build a system that detects specific sounds using the onboard microphone. By collecting data, training models, and deploying them onto the microcontroller, students will experience the complete machine learning pipeline, from data collection to real-time inference.