首页> 外国专利> Temperature compensation and leakage compensation for memory cells of analog neural memory systems used in deep learning neural networks

Temperature compensation and leakage compensation for memory cells of analog neural memory systems used in deep learning neural networks

机译:深度学习神经网络中使用的模拟神经记忆系统存储器的温度补偿和泄漏补偿

摘要

Numerous embodiments are disclosed to provide temperature compensation and leakage compensation for analog neuromorphic memory systems used in deep learning neural networks. Embodiments for providing temperature compensation implement careful or continuous adaptive tilt compensation and renormalization for selected memory cells of a device, reference memory cell, or memory system. An embodiment for providing leak compensation in a memory cell of a memory system is an adaptive erase gate coupling or application of a negative bias to the control gate terminal, a negative bias to the word line terminal, or a bias to the source line terminal. Implement. [Selection diagram] FIG. 6
机译:公开了许多实施方案,以提供深度学习神经网络中使用的模拟神经晶体记忆系统的温度补偿和泄漏补偿。 用于提供温度补偿的实施例为设备,参考存储器单元或存储器系统的所选存储器单元实现仔细或连续的自适应倾斜补偿和重整化。 用于在存储器系统的存储器单元中提供泄漏补偿的实施例是对控制栅极终端的反向栅极耦合或施加负偏压,对字线终端的负偏压,或源极线终端的偏置。 实施。 [选择图]图。 6.

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号