[Workshop] tinyML Talks: AIML SoC for Ultra-Low-Power Mobile and IoT devices

Published: July 22, 2020 by SingularityKChen (Last updated: July 25, 2020 )

• Categories:
• Workshop 11

• Tags:
• Dequantization 1
• Entropy 2
• tinyML 3

• tinyML Talks: AIML SoC for Ultra-Low-Power Mobile and IoT devices
  • Background
  • Acceleration
    • Computation Optimization
    • Memory Optimization

tinyML Talks: AIML SoC for Ultra-Low-Power Mobile and IoT devices

Background

Acceleration

Computation Optimization

Efficient Control Unit: Address Generation HW Unit

Highest Time Utilization: Pipeline Architecture

Memory Optimization

Dequantization

Entropy Compression

Pooling on the Fly