This paper reports a comprehensive study on the impacts of temperature-change, process variation, flicker noise and device aging on the inference accuracy of pre-trained all-ferroelectric (FE) FinFET deep neural networks. Multiple-level-cell (MLC) operation with a novel adaptive-program-and-read algorithm with 100ns write pulse has been experimentally demonstrated in 5 nm thick hafnium zirconium oxide (HZO)-based FE-FinFET. With pre-trained neural network (NN) with 97.5% inference accuracy on MNIST dataset as baseline, device to device variation is shown to have negligible impact. Flicker noise characterization at various bias conditions depicts that drain current fluctuation is less than 0.7% with virtually no inference accuracy degradation. The conductance drift of a programmed cell, as an aftermath of temperature change, was captured by a compact model over a wide range of gate biases. Despite significant inference accuracy degradation at 233K for a NN trained at 300K, gate bias optimization for recovering the accuracy is demonstrated. Endurance above 10$^8$ cycles and extrapolated retention above 10 years are shown, which paves the way for edge device artificial intelligence with FE-FinFETs.
翻译:本文报告了对温度变化、工艺变异、闪烁噪音和装置影响的全面研究,这些影响是在经过预先训练的全热电网(FE)FinFET深神经网络的推断精度方面出现的。多级细胞(MLC)的操作具有一种新型的适应-方案和读算算法,具有100ns写脉冲100ns的写式脉冲。在5纳米厚厚的氢氟氧化 ⁇ (HZO)以FE-FinFET为基地的FE-FinFET中,实验地展示了程序化细胞的导向移动。尽管在300K培训的NNNT中,233K的精度明显下降,但显示设备变异的装置影响微不足道。各种偏差条件下的闪亮噪音特征表明,目前的抽水波动不到0.7%,而几乎没有推断精度降解。在温度变化之后,一个程序化的细胞的行走势在广泛的门偏差上被一个压缩模型所捕捉到。尽管在300KNIST(N)训练的精确度为95%,但用于恢复精确度的门偏差优化的装置显示为10美元以上。