Model Pruning & Quantization in TinyML | Seminar Lecture 2 (Practical Session) скачать в хорошем качестве

Model Pruning & Quantization in TinyML | Seminar Lecture 2 (Practical Session)

This video is a recording of the second session from our TinyML seminar at Mälardalen University (MDU), focused on model pruning and quantization for embedded and edge AI systems. ⚠️ Note: The recording did not capture the very beginning of the session due to a late start. However, the core technical content, explanations, and live demonstrations are fully intact and provide strong practical value. In this session, we cover: • Motivation for model pruning in TinyML • Structured vs unstructured pruning concepts • Accuracy–efficiency trade-offs • Practical pruning workflows • Live notebook demonstrations on quantization and pruning • Combining pruning and quantization for efficient deployment The session emphasizes hands-on understanding, showing how compression techniques can be applied in practice to reduce model size, computation, and energy consumption for resource-constrained devices. This lecture is intended for students, researchers, and engineers working in embedded systems, edge AI, and TinyML. 📂 Seminar materials and notebooks: https://github.com/HERO-MDH/TinyML-Se... ▶️ Part of the TinyML Seminar @ MDU playlist.