NV-BNN: An accurate deep convolutional neural network based on binary STT-MRAM for adaptive AI edge
Binary STT-MRAM is a highly anticipated embedded nonvolatile memory technology in
advanced logic nodes< 28 nm. How to enable its in-memory computing (IMC) capability is
critical for enhancing AI Edge. Based on the soon-available STT-MRAM, we report the first
binary deep convolutional neural network (NV-BNN) capable of both local and remote
learning. Exploiting intrinsic cumulative switching probability, accurate online training of
CIFAR-10 color images (~ 90%) is realized using a relaxed endurance spec (switching≤ 20 …
advanced logic nodes< 28 nm. How to enable its in-memory computing (IMC) capability is
critical for enhancing AI Edge. Based on the soon-available STT-MRAM, we report the first
binary deep convolutional neural network (NV-BNN) capable of both local and remote
learning. Exploiting intrinsic cumulative switching probability, accurate online training of
CIFAR-10 color images (~ 90%) is realized using a relaxed endurance spec (switching≤ 20 …
以上显示的是最相近的搜索结果。 查看全部搜索结果