15.2 A 28nm 64Kb Inference-Training Two-Way Transpose Multibit 6T SRAM Compute-in-Memory Macro for AI Edge Chips.
Jian-Wei SuXin SiYen-Chi ChouTing-Wei ChangWei-Hsing HuangYung-Ning TuRuhui LiuPei-Jung LuTa-Wei LiuJing-Hong WangZhixiao ZhangHongwu JiangShanshi HuangChung-Chuan LoRen-Shuo LiuChih-Cheng HsiehKea-Tiong TangShyh-Shyuan SheuSih-Han LiHeng-Yuan LeeShih-Chieh ChangShimeng YuMeng-Fan ChangPublished in: ISSCC (2020)
Keyphrases
- artificial intelligence
- knowledge base
- expert systems
- structured prediction
- power consumption
- case based reasoning
- random access
- probabilistic inference
- machine learning
- knowledge representation
- supervised learning
- ai systems
- high end
- random access memory
- computer systems
- knowledge based systems
- intelligent systems
- memory requirements
- training set
- semantic role labeling
- digital signal processors
- dynamic random access memory