Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 21 hours
Course Outline
Introduction to TinyML and Embedded AI
- Key aspects of deploying TinyML models
- Limitations within microcontroller settings
- Review of embedded AI toolchains
Foundations of Model Optimization
- Recognizing computational bottlenecks
- Spotting operations that heavily consume memory
- Establishing baseline performance profiles
Quantization Methods
- Strategies for post-training quantization
- Quantization-aware training processes
- Assessing the trade-off between accuracy and resource usage
Pruning and Compression
- Techniques for structured and unstructured pruning
- Implementing weight sharing and model sparsity
- Algorithms for compressing models for lightweight inference
Hardware-Specific Optimization
- Model deployment on ARM Cortex-M architectures
- Optimizing for DSP and accelerator extensions
- Considerations for memory mapping and dataflow
Benchmarking and Validation
- Analyzing latency and throughput
- Measuring power and energy consumption
- Testing for accuracy and robustness
Deployment Processes and Tools
- Utilizing TensorFlow Lite Micro for embedded applications
- Incorporating TinyML models into Edge Impulse pipelines
- Testing and debugging on physical hardware
Advanced Optimization Tactics
- Applying neural architecture search to TinyML
- Combining quantization and pruning techniques
- Using model distillation for embedded inference
Conclusion and Future Directions
Requirements
- A solid grasp of machine learning processes
- Hands-on experience with embedded systems or microcontroller-based projects
- Proficiency in Python programming
Intended Participants
- Researchers in artificial intelligence
- Engineers specializing in embedded ML
- Specialists focused on inference systems with limited resources
