Part XII: Edge, Embedded, Streaming, and Federated Sensor AI
Chapter 59  [C][A]

Edge AI Fundamentals and Model Optimization

Placeholder chapter page. Sections are produced by the book-skills pipeline.

Sections

  1. 59.1 Why sensory AI runs on-device
  2. 59.2 CPU, GPU, NPU, DSP, MCU tradeoffs
  3. 59.3 Memory, power, and thermal limits
  4. 59.4 Quantization (int8, int4, and hardware-native 2/4-bit)
  5. 59.5 Pruning and sparsity
  6. 59.6 Hardware-aware NAS
  7. 59.7 Edge-cloud partitioning and the deployment lifecycle

Lab 59

quantize and prune a sensor classifier for edge inference; measure latency/size/accuracy tradeoffs.