Flash-based content addressable memory with L2 distance for memory-augmented neural network
Haozhang Yang,
Peng Huang,
Ruiyi Li,
Nan Tang,
Yizhou Zhang,
Zheng Zhou,
Lifeng Liu,
Xiaoyan Liu,
Jinfeng Kang
Affiliations
Haozhang Yang
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Peng Huang
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China; Corresponding author
Ruiyi Li
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Nan Tang
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Yizhou Zhang
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Zheng Zhou
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Lifeng Liu
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Xiaoyan Liu
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Jinfeng Kang
School of Integrated Circuits, Peking University, Beijing 100871, China; Beijing Advanced Innovation Center for Integrated Circuits, Beijing 100871, China
Summary: Memory-augmented neural network (MANN) has received increasing attention as a promising approach to achieve lifelong on-device learning, of which implementation of the explicit memory is vital. Content addressable memory (CAM) has been designed to accelerate the explicit memory by harnessing the in-memory-computing capability. In this work, a CAM cell with quadratic code is proposed, and a 1Mb Flash-based multi-bit CAM chip capable of computing Euclidean (L2) distance is fabricated. Compared with ternary CAM, the latency and energy are significantly reduced by 5.3- and 46.6-fold, respectively, for the MANN on Omniglot dataset. Besides, the recognition accuracy has slight degradation (<1%) even after baking for 105 s at 200°C, demonstrating the robustness to environmental disturbance. Performance evaluation indicates a reduction of 471-fold in latency and 1267-fold in energy compared with GPU for search operation. The proposed robust and energy-efficient CAM provides a promising solution to implement lifelong on-device machine intelligence.
