A 12-nm 0.62-1.61 mW Ultra-Low Power Digital CIM-based Deep-Learning System for End-to-End Always-on Vision.
En-Jui ChangCheng-Xin XueChetan DeshpandeGajanan JedheJenwei LiangChih-Chung ChengHung-Wei LinChia-Da LeeSushil KumarKim Soon JwayZijie GuoRitesh GargAllen-Cl LuChien-Hung LinMeng-Han HsiehTsung-Yao LinChih-Cheng ChenPublished in: VLSI Technology and Circuits (2023)
Keyphrases
- probabilistic model
- end to end
- deep learning
- ultra low power
- machine learning
- unsupervised learning
- power consumption
- low power
- computer vision
- unsupervised feature learning
- congestion control
- weakly supervised
- real time
- image processing
- transport layer
- mental models
- application layer
- image segmentation
- feature selection