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 attributes of TinyML model deployment
- Limitations specific to microcontroller environments
- Overview of tools available for embedded AI
Foundations of Model Optimization
- Recognizing computational bottlenecks
- Detecting operations that consume significant memory
- Establishing baseline performance metrics
Quantization Methods
- Strategies for post-training quantization
- Quantization-aware training approaches
- Assessing the impact of accuracy versus resource usage
Pruning and Compression
- Techniques for structured and unstructured pruning
- Implementing weight sharing and model sparsity
- Compression algorithms designed for lightweight inference
Hardware-Specific Optimization
- Running models on ARM Cortex-M architectures
- Optimizing for DSP and accelerator capabilities
- Considerations for memory mapping and dataflow
Benchmarking and Verification
- Analyzing latency and throughput
- Measuring power and energy usage
- Testing for accuracy and robustness
Deployment Processes and Utilities
- Leveraging TensorFlow Lite Micro for embedded systems
- Connecting TinyML models with Edge Impulse workflows
- Testing and troubleshooting on physical hardware
Advanced Optimization Approaches
- Applying neural architecture search to TinyML
- Combining quantization and pruning strategies
- Using model distillation for embedded inference
Conclusions and Future Directions
Requirements
- Knowledge of machine learning workflows
- Experience with embedded systems or microcontroller-based projects
- Proficiency in Python programming
Target Audience
- AI researchers
- Embedded ML engineers
- Professionals developing inference systems with resource constraints
