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
- Defining characteristics of TinyML model deployment
- Specific constraints within microcontroller environments
- Overview of available embedded AI toolchains
Foundations of Model Optimization
- Identifying computational bottlenecks
- Recognizing memory-intensive operations
- Conducting baseline performance profiling
Quantization Techniques
- Strategies for post-training quantization
- Implementing quantization-aware training
- Balancing accuracy against resource consumption
Pruning and Compression
- Methods for structured and unstructured pruning
- Utilizing weight sharing and model sparsity
- Applying compression algorithms for lightweight inference
Hardware-Aware Optimization
- Deploying models on ARM Cortex-M systems
- Leveraging DSP and accelerator extensions for optimization
- Considerations for memory mapping and dataflow
Benchmarking and Validation
- Analyzing latency and throughput
- Measuring power and energy consumption
- Testing for accuracy and robustness
Deployment Workflows and Tools
- Employing TensorFlow Lite Micro for embedded deployment
- Integrating TinyML models with Edge Impulse pipelines
- Testing and debugging on physical hardware
Advanced Optimization Strategies
- Applying neural architecture search to TinyML
- Combining quantization and pruning approaches
- Using model distillation for embedded inference
Summary and Next Steps
Requirements
- A solid grasp of machine learning workflows
- Experience in embedded systems or microcontroller-based development
- Proficiency in Python programming
Target Audience
- AI researchers
- Embedded ML engineers
- Professionals specializing in resource-constrained inference systems
