28-nm FD-SOI Dual-Port SRAM with MSB-Based Inversion Logic for Low-Power Deep Learning.
Haruki MoriShintaro IzumiHiroshi KawaguchiMasahiko YoshimotoPublished in: ICECS (2018)
Keyphrases
- low power
- deep learning
- cmos technology
- silicon on insulator
- power consumption
- logic circuits
- high speed
- low cost
- power reduction
- delay insensitive
- nm technology
- unsupervised learning
- embedded dram
- unsupervised feature learning
- low voltage
- random access memory
- power dissipation
- machine learning
- digital signal processing
- low power consumption
- mixed signal
- mental models
- power management
- weakly supervised
- image sensor
- supervised learning
- video sequences
- relation extraction
- pattern recognition
- parallel processing
- real time