A system-level exploration of binary neural network accelerators with monolithic 3D based compute-in-memory SRAM

JH Choi, YH Gong, SW Chung - Electronics, 2021 - mdpi.com
Binary neural networks (BNNs) are adequate for energy-constrained embedded systems
thanks to binarized parameters. Several researchers have proposed the compute-in-
memory (CiM) SRAMs for XNOR-and-accumulation computations (XACs) in BNNs by
adding additional transistors to the conventional 6T SRAM, which reduce the latency and
energy of the data movements. However, due to the additional transistors, the CiM SRAMs
suffer from larger area and longer wires than the conventional 6T SRAMs. Meanwhile …

[PDF][PDF] A System-Level Exploration of Binary Neural Network Accelerators with Monolithic 3D Based Compute-in-Memory SRAM. Electronics 2021, 10, 623

JH Choi, YH Gong, SW Chung - 2021 - academia.edu
Binary neural networks (BNNs) are adequate for energy-constrained embedded systems
thanks to binarized parameters. Several researchers have proposed the compute-in-
memory (CiM) SRAMs for XNOR-and-accumulation computations (XACs) in BNNs by
adding additional transistors to the conventional 6T SRAM, which reduce the latency and
energy of the data movements. However, due to the additional transistors, the CiM SRAMs
suffer from larger area and longer wires than the conventional 6T SRAMs. Meanwhile …
以上显示的是最相近的搜索结果。 查看全部搜索结果