Toward Data-Driven Applications with AIoT Devices

Explore data-driven applications and AIoT devices in this 32-minute conference talk from tinyML Asia 2020. Dive into model reduction techniques like network architecture search and adaptive quantization to optimize storage and computing efficiency while maintaining accuracy. Learn about a meta-learning approach for enhancing inference accuracy in diverse environments. Examine real-life examples in image and video services, and gain insights into future AIoT research directions. Discover how local intelligence solutions in IoT devices can improve data security and real-time performance under bandwidth constraints.

Introduction
Outline
AI Solution
New Business Opportunity
Automatic Gauge Reader
Weightable ECG Devices
Entry Control and Video Surveillance
Guidelines
Technical Issues
Fine Tuning
Neural Scale
Uniform Scanning
Optimization
Deployment Issues
Transductive Inference
Network Architecture
Distance minimization
Feasibility
Demonstration
Summary
Questions
Sponsors