CN113964121A - 一种跨导可变场效应晶体管阵列及应用 - Google Patents
一种跨导可变场效应晶体管阵列及应用 Download PDFInfo
- Publication number
- CN113964121A CN113964121A CN202111208889.9A CN202111208889A CN113964121A CN 113964121 A CN113964121 A CN 113964121A CN 202111208889 A CN202111208889 A CN 202111208889A CN 113964121 A CN113964121 A CN 113964121A
- Authority
- CN
- China
- Prior art keywords
- tantalum
- field effect
- metal
- array
- variable field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005669 field effect Effects 0.000 title claims abstract description 32
- 239000004065 semiconductor Substances 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 10
- 238000003860 storage Methods 0.000 claims abstract description 6
- 238000013507 mapping Methods 0.000 claims abstract description 4
- 239000010410 layer Substances 0.000 claims description 37
- 229910052751 metal Inorganic materials 0.000 claims description 28
- 239000002184 metal Substances 0.000 claims description 28
- 229910052715 tantalum Inorganic materials 0.000 claims description 26
- GUVRBAGPIYLISA-UHFFFAOYSA-N tantalum atom Chemical compound [Ta] GUVRBAGPIYLISA-UHFFFAOYSA-N 0.000 claims description 25
- 229910044991 metal oxide Inorganic materials 0.000 claims description 19
- 150000004706 metal oxides Chemical class 0.000 claims description 19
- 239000010936 titanium Substances 0.000 claims description 13
- 150000001875 compounds Chemical class 0.000 claims description 9
- 229910052719 titanium Inorganic materials 0.000 claims description 8
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 claims description 6
- 239000002131 composite material Substances 0.000 claims description 6
- 229910052721 tungsten Inorganic materials 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 5
- 239000002356 single layer Substances 0.000 claims description 5
- 229910001936 tantalum oxide Inorganic materials 0.000 claims description 5
- 229910004166 TaN Inorganic materials 0.000 claims description 4
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 claims description 4
- 238000003491 array Methods 0.000 claims description 4
- 229910052802 copper Inorganic materials 0.000 claims description 4
- GKOZUEZYRPOHIO-UHFFFAOYSA-N iridium atom Chemical compound [Ir] GKOZUEZYRPOHIO-UHFFFAOYSA-N 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000002161 passivation Methods 0.000 claims description 4
- 229910052718 tin Inorganic materials 0.000 claims description 4
- 229910003070 TaOx Inorganic materials 0.000 claims description 3
- 229910052741 iridium Inorganic materials 0.000 claims description 3
- 229910052697 platinum Inorganic materials 0.000 claims description 3
- 229910017083 AlN Inorganic materials 0.000 claims description 2
- 229910017107 AlOx Inorganic materials 0.000 claims description 2
- 229910003134 ZrOx Inorganic materials 0.000 claims description 2
- 229910052782 aluminium Inorganic materials 0.000 claims description 2
- 150000002739 metals Chemical class 0.000 claims description 2
- 229910000314 transition metal oxide Inorganic materials 0.000 claims description 2
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 claims description 2
- 239000010937 tungsten Substances 0.000 claims description 2
- 229910052735 hafnium Inorganic materials 0.000 claims 2
- VBJZVLUMGGDVMO-UHFFFAOYSA-N hafnium atom Chemical compound [Hf] VBJZVLUMGGDVMO-UHFFFAOYSA-N 0.000 claims 2
- 239000010408 film Substances 0.000 claims 1
- -1 respectivelyx Inorganic materials 0.000 claims 1
- 239000010409 thin film Substances 0.000 claims 1
- OGIDPMRJRNCKJF-UHFFFAOYSA-N titanium oxide Inorganic materials [Ti]=O OGIDPMRJRNCKJF-UHFFFAOYSA-N 0.000 claims 1
- 230000004913 activation Effects 0.000 abstract description 9
- 238000013473 artificial intelligence Methods 0.000 abstract description 7
- 230000009466 transformation Effects 0.000 abstract description 7
- 230000000295 complement effect Effects 0.000 abstract description 6
- 238000013461 design Methods 0.000 abstract description 6
- 210000002569 neuron Anatomy 0.000 abstract description 6
- 238000013528 artificial neural network Methods 0.000 abstract description 5
- 230000002093 peripheral effect Effects 0.000 abstract description 4
- 238000005240 physical vapour deposition Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000005530 etching Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000000059 patterning Methods 0.000 description 3
- 238000000206 photolithography Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000007772 electrode material Substances 0.000 description 2
- 238000005566 electron beam evaporation Methods 0.000 description 2
- WIHZLLGSGQNAGK-UHFFFAOYSA-N hafnium(4+);oxygen(2-) Chemical class [O-2].[O-2].[Hf+4] WIHZLLGSGQNAGK-UHFFFAOYSA-N 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 239000011185 multilayer composite material Substances 0.000 description 2
- BPUBBGLMJRNUCC-UHFFFAOYSA-N oxygen(2-);tantalum(5+) Chemical compound [O-2].[O-2].[O-2].[O-2].[O-2].[Ta+5].[Ta+5] BPUBBGLMJRNUCC-UHFFFAOYSA-N 0.000 description 2
- 238000000623 plasma-assisted chemical vapour deposition Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 210000000225 synapse Anatomy 0.000 description 2
- 229910003087 TiOx Inorganic materials 0.000 description 1
- 238000000137 annealing Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 210000001787 dendrite Anatomy 0.000 description 1
- 239000003989 dielectric material Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 229910000457 iridium oxide Inorganic materials 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000001259 photo etching Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000946 synaptic effect Effects 0.000 description 1
- HLLICFJUWSZHRJ-UHFFFAOYSA-N tioxidazole Chemical compound CCCOC1=CC=C2N=C(NC(=O)OC)SC2=C1 HLLICFJUWSZHRJ-UHFFFAOYSA-N 0.000 description 1
- 229910052720 vanadium Inorganic materials 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
- H01L27/02—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having potential barriers; including integrated passive circuit elements having potential barriers
- H01L27/04—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having potential barriers; including integrated passive circuit elements having potential barriers the substrate being a semiconductor body
- H01L27/08—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having potential barriers; including integrated passive circuit elements having potential barriers the substrate being a semiconductor body including only semiconductor components of a single kind
- H01L27/085—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components specially adapted for rectifying, oscillating, amplifying or switching and having potential barriers; including integrated passive circuit elements having potential barriers the substrate being a semiconductor body including only semiconductor components of a single kind including field-effect components only
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Neurology (AREA)
- Computer Hardware Design (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Semiconductor Memories (AREA)
Abstract
本发明公布了一种适用于树突网络硬件的跨导可变场效应晶体管阵列及应用,属于半导体集成电路技术领域。本发明基于单个跨导可变场效应晶体管实现存储变量与两个输入变量的三元素乘法,并基于互补器件阵列实现了树突网络核心算法的映射。相比于利用神经元激活电路实现非线性变换的传统神经网络硬件,本发明利用器件的本征非线性实现非线性变换,有效降低了设计复杂性,优化了系统外围电路的面积和功耗,对高性能人工智能计算系统的设计具有重要意义。
Description
技术领域
本发明属于半导体(semiconductor)、人工智能(artificial intelligence)和互补型金属氧化物半导体(CMOS)混合集成电路技术领域,具体涉及一种适用于树突网络硬件的跨导可变场效应晶体管阵列及应用。
背景技术
人工智能的概念源起于20世纪50年代。历经半个多世纪的发展,进入21世纪后,人工智能在机器视觉,语音识别等领域都取得了巨大的成功。互联网蓬勃发展带来的爆炸式信息增长在促进人工智能的发展同时,也对硬件的计算能力提出了新的要求。近年来,以存储器阵列为核心的存算一体计算系统受到了广泛的关注。存储器阵列在实现存储信息的功能的同时还兼顾对信息的处理,在一定程度上克服了传统冯诺依曼计算系统存在的“存储墙”问题。
以两端器件阵列为核心的神经形态计算系统已多有报道。由于器件电导值可连续调节,因此基于金属-介质层-金属结构的两端阻变器件被广泛应用于模拟突触的功能。在人工神经网络(ANN)中,突触网络实现线性变换功能,而非线性激活功能则由神经元实现。因此,阻变存储器阵列必须搭配相应的神经元激活电路才能实现网络的非线性变换功能。由于网络的每一层都需要激活电路,当网络深度增加时,神经元激活电路的实现将耗费大量外围电路,不仅造成了系统功耗、面积的增加,还提升了系统的设计难度,阻碍了大规模计算系统的开发。此外,现有单个器件难以实现多元变量的相乘,需要多个晶体管组成的CMOS电路来实现多元乘法,有着较大的硬件开销。
发明内容
为克服现有神经形态计算系统中外围激活函数电路面积过大和多元计算硬件开销大的问题,本发明提出了一种适用于树突网络硬件的跨导可变场效应晶体管阵列,利用器件本征方程实现三元素乘法计算,最终实现输入到输出的非线性变换,免除了层与层之间额外的非线性激活电路,降低了系统设计的复杂度。同时,该器件以阵列具有可后端低温集成的潜力,具有三维高密度集成的潜力。
为解决上述技术问题,本发明采用的技术方案如下:
一种跨导可变场效应晶体管阵列,其特征在于,由多个跨导可变场效应晶体管器件连接形成,所述跨导可变场效应晶体管包括金属栅电极、金属氧化物材料介质层、化合物半导体有源层、金属源、漏电极和钝化层,所述跨导可变场效应晶体管的三元乘法实现方法,主要包括输入输出映射方式、信息存储方式、以及利用器件本征方程实现计算的方法。场效应晶体管的跨导gm由两个因子相乘表示,即gm=w×Vd,其中系数w表示栅极对沟道的控制能力,Vd代表漏极与源极之间的电势差,分别控制w和Vd即可控制器件的跨导。通过改变金属氧化物介质层与沟道有源区界面处存储的电荷数量,可以改变栅极对沟道的控制能力w,从而改变器件跨导gm。器件沟道电流最终由栅极电压(Vg)与跨导(gm)的乘积决定,即由w,Vd,Vg三者乘积决定。因此,单个器件即可完成其存储信息(w)与两输入信息(Vd、Vg)之间的三元乘积计算(Id=w×Vd×Vg)。由于输出(Id)是输入(Vd,Vg)的二次函数,因此单个器件实现了输入到输出的非线性变换;在阵列中,同一行器件的漏电极由字线(Word line,WL)相连接而源电极由源线(Source line,SL)相连接,同一列器件的栅极由位线(Bit line,BL)相连接。单个器件完成其存储信息(wij)与两输入信息(VWL j、VBL i)之间的三元乘积计算,电流汇入源线实现电流加和。当源线电位被钳制为0V时,每一条源线上的电流可由下列公式表示。
可以看到,存储矩阵W先与输入向量VBL做内积,得到的结果再与另一输入向量VWL做元素积,最终得到源线的电流向量ISL。上述运算的矩阵形式如下:
ISL=WVBL⊙VWL
树突网络核心迭代公式如下:
Ai+1=WAi⊙A0
因此,只需将ISL、W、VBL和VWL与网络中Ai+1、W、Ai和A0做一一映射,即可实现单层树突网络的运算。当多个阵列通过接口电路级联后,即可将整个树突网络映射至硬件,加速其运算。且正负权重由互补跨导可变场效应晶体管阵列实现,正权重阵列与负权重阵列共享输入,分别得到源线电流ISL +和ISL -,两者相减得到最终的源线电流ISL。通过互补阵列电流相减的模式,正、负权值阵列对应器件电流中的一次相、零次项成分之间将相互抵消,有效缓解了器件非理想效应对输出结果的影响。
优选地,所述金属电极(包括栅电极、源电极以及漏电极)材料为导体材料,包括Ti、TiN、TaN、Ta、Al、AlN、W、Cu、Pt等;
优选地,所述金属氧化物介质层为由单层或多层复合材料薄膜组成,包括金属钽和金属氧化物的复合材料,包括钽和钽的氧化物(Ta/TaOx)、钽和铪的氧化物(Ta/HfOx),或是金属钽、其它金属和金属氧化物的复合材料,包括钽和钛和钽的氧化物(Ta/Ti/TaOx)、钽和钛和铪的氧化物(Ta/Ti/HfOx)、钽和铱和钽的氧化物(Ta/Ir/TaOx)、钽和钨和钽的氧化物(Ta/W/TaOx)、钽和铱和钛的氧化物(Ta/Ir/TiOx),在以上金属钽和金属氧化物的复合材料的金属氧化物端可以是多种金属材料,包括Cu、Ti、Ta、W、Pt、TiN、TaN、TiOx、TaOx、WOx、HfOx、AlOx、ZrOx等,形成金属/N层过渡金属氧化物/金属结构,N≥1
优选地,所述化合物半导体有源层为由单层或多层复合材料薄膜组成,包括ZnO,InGaZnO,ITO,VOx、NbOx等。
优选地,所述金属氧化物介质层薄膜和化合物半导体有源层材料薄膜厚度为5nm-1000nm。
本发明提出了面向树突网络硬件的跨导可变场效应晶体管阵列。基于单个跨导可变场效应晶体管实现存储变量与两个输入变量的三元素乘法,并基于互补器件阵列实现了树突网络核心算法的映射。相比于利用神经元激活电路实现非线性变换的传统神经网络硬件,本发明利用器件的本征非线性实现非线性变换,有效降低了设计复杂性,优化了系统外围电路的面积和功耗,对高性能人工智能计算系统的设计具有重要意义。
附图说明
图1是本发明跨导可变场效应晶体管的结构示意图;
图2是本发明跨导可变场效应晶体管通过调节w调节跨导的示意图;
图3是本发明跨导可变场效应晶体管改变漏极电压以调节器件跨导的示意图;
图4是本发明跨导可变场效应晶体管阵列的结构示意图;
图5是本发明正权重阵列与负权重阵列共享输入示意图;
图6是本发明跨导可变场效应晶体管阵列构成的树突网络硬件的示意图;图中4×4个圆圈代表图4所示的跨导可变场效应晶体管阵列;
图7是本发明实施例树突网络对MNIST数据集的识别精度随权重量化状态数示意图。
具体实施方式
为使本发明的上述特征和优点能更明显易懂,下面结合附图和具体实施例,对本发明进行进一步描述。
本发明提供一种基于跨导可变场效应晶体管阵列的树突网络硬件,利用三端器件自身的非线性,实现输入到输出的非线性变换,免除了层与层之间额外的非线性激活电路,具体如下:
本发明基于陷阱电荷的跨导可变场效应晶体管,包括金属栅电极、金属氧化物材料介质层、化合物半导体有源层、金属源、漏电极和钝化层。
其制备方法包括如下步骤:
1)利用物理气相淀积(PVD)、电子束蒸发等方法形成金属栅电极材料;
2)利用光刻和刻蚀图形化金属栅电极;
3)利用物理气相淀积(PVD)方法形成金属氧化物介质材料层,作为背栅介质;
4)利用物理气相淀积(PVD)方法形成化合物半导体沟道有源层;
5)利用光刻和刻蚀图形化器件有源区;
6)退火调整沟道有源区组分;
7)利用物理气相淀积(PVD)、电子束蒸发等方法形成金属源、漏电极材料;
8)利用光刻和刻蚀图形化金属源、漏电极;
9)利用利用物理气相淀积(PVD)、等离子增强化学气相沉积(PECVD)等方法形成钝化层;
10)利用光刻和刻蚀图形化接触孔。
以电极为Ti/Pt,金属氧化物栅介质层和化合物半导体沟道有源层分别为TaOx和InGaZnO的三端非线性器件为例,可以通过调节金属氧化物介质层与沟道有源区界面处存储的电荷数量,可以改变栅极对沟道的控制能力w,从而改变器件跨导。通过调节w调节跨导的过程如图2所示;改变漏极电压以调节器件跨导的过程如图3所示;跨导可变场效应晶体管阵列中,同一行器件的漏电极由字线(Word line,WL)相连接而源电极由源线(Sourceline,SL)相连接,同一列器件的栅极由位线(Bit line,BL)相连接。阵列结构如图4所示;正负权重由互补器件阵列实现,正权重阵列与负权重阵列共享输入,分别得到源线电流ISL +和ISL -,两者相减得到最终的源线电流ISL,如图5所示;树突网络硬件可以分为三层,分别为突触层、树突层以及神经元层,其中树突层由跨导可变场效应晶体管阵列实现,如图6所示;树突网络对MNIST数据集的识别精度随权重量化状态数目上升而上升,对于64×64规模的两层树突网络,在量化至3bit后,可以达到95.67%的识别准确率,与传统神经网络相当,如图7所示。
最后需要注意的是,公布实施例的目的在于帮助进一步理解本发明,但是本领域的技术人员可以理解:在不脱离本发明及所附的权利要求的精神和范围内,各种替换和修改都是可能的。因此,本发明不应局限于实施例所公开的内容,本发明要求保护的范围以权利要求书界定的范围为准。
Claims (9)
1.一种跨导可变场效应晶体管阵列,其特征在于,由多个跨导可变场效应晶体管连接形成,所述跨导可变场效应晶体管包括金属栅电极、金属氧化物材料介质层、化合物半导体有源层、金属源、漏电极和钝化层,同一行跨导可变场效应晶体管的漏电极由字线相连接;同一行跨导可变场效应晶体管的源电极由源线相连接,同一列跨导可变场效应晶体管的栅极由位线相连接;单个跨导可变场效应晶体管完成其存储信息(wij)与两输入信息(VWL j、VBL 1)之间的三元乘积计算,电流汇入源线实现电流加和。
2.如权利要求1所述的跨导可变场效应晶体管阵列,其特征在于,所述金属栅电极、金属源、漏电极采用Ti、TiN、TaN、Ta、Al、AlN、W、Cu或Pt。
3.如权利要求1所述的跨导可变场效应晶体管阵列,其特征在于,所述金属氧化物介质层由单层或多层复合材料薄膜组成,所述复合材料为金属钽和金属氧化物,或是金属钽、其它金属和金属氧化物。
4.如权利要求3所述的跨导可变场效应晶体管阵列,其特征在于,所述金属和金属氧化物分别为Cu、Ti、Ta、W、Pt、TiN、TaN、TiOx、TaOx、WOx、HfOx、AlOx或ZrOx,或者上述金属形成金属/N层过渡金属氧化物/金属结构,N≥1。
5.如权利要求3所述的跨导可变场效应晶体管阵列,其特征在于,所述复合材料为包括钽和钽的氧化物、钽和铪的氧化物;或是钽和钛和钽的氧化物、钽和钛和铪的氧化物、钽和铱和钽的氧化物、钽和钨和钽的氧化物、钽和铱和钛的氧化物。
6.如权利要求1所述的跨导可变场效应晶体管阵列,其特征在于,所述化合物半导体有源层为由单层或多层ZnO、InGaZnO、ITO、VOx或NbOx。
7.如权利要求1所述的跨导可变场效应晶体管阵列,其特征在于,所述金属氧化物介质层薄膜和化合物半导体有源层厚度分别为5nm-1000nm。
9.如权利要求8所述的运算方法,其特征在于,所述两个跨导可变场效应晶体管阵列分别作为正权重阵列与负权重阵列共享输入,分别得到源线电流ISL +和ISL -,两者相减得到最终的源线电流ISL。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111208889.9A CN113964121A (zh) | 2021-10-18 | 2021-10-18 | 一种跨导可变场效应晶体管阵列及应用 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111208889.9A CN113964121A (zh) | 2021-10-18 | 2021-10-18 | 一种跨导可变场效应晶体管阵列及应用 |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113964121A true CN113964121A (zh) | 2022-01-21 |
Family
ID=79464898
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111208889.9A Pending CN113964121A (zh) | 2021-10-18 | 2021-10-18 | 一种跨导可变场效应晶体管阵列及应用 |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113964121A (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024092406A1 (zh) * | 2022-10-31 | 2024-05-10 | 北京大学 | 互补光电晶体管像素单元、感算阵列结构及其操作方法 |
-
2021
- 2021-10-18 CN CN202111208889.9A patent/CN113964121A/zh active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024092406A1 (zh) * | 2022-10-31 | 2024-05-10 | 北京大学 | 互补光电晶体管像素单元、感算阵列结构及其操作方法 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Brain-inspired computing with memristors: Challenges in devices, circuits, and systems | |
Lim et al. | Adaptive learning rule for hardware-based deep neural networks using electronic synapse devices | |
Prezioso et al. | Training and operation of an integrated neuromorphic network based on metal-oxide memristors | |
US10164179B2 (en) | Memristive device based on alkali-doping of transitional metal oxides | |
Moon et al. | Improved conductance linearity and conductance ratio of 1T2R synapse device for neuromorphic systems | |
Liu et al. | Optimization of non-linear conductance modulation based on metal oxide memristors | |
Mannocci et al. | In-memory computing with emerging memory devices: Status and outlook | |
WO2020226740A1 (en) | Transistorless all-memristor neuromorphic circuits for in-memory computing | |
CN111680792A (zh) | 激活函数电路、忆阻神经网络及忆阻神经网络的控制方法 | |
AU2021395683A1 (en) | Non-volatile analog resistive memory cells implementing ferroelectric select transistors | |
Wang et al. | Self-activation neural network based on self-selective memory device with rectified multilevel states | |
US20210327502A1 (en) | Analog computing architecture for four terminal memory devices | |
CN113964121A (zh) | 一种跨导可变场效应晶体管阵列及应用 | |
US11121259B2 (en) | Metal-oxide-based neuromorphic device | |
CN115768248A (zh) | 氧化铪基铁电ftj忆阻器及其制备方法、芯片 | |
Li et al. | Monolithic three-dimensional integration of RRAM-based hybrid memory architecture for one-shot learning | |
Wang et al. | Emerging memristive devices for brain-inspired computing and artificial perception | |
Kim et al. | Nano-crossbar weighted memristor-based convolution neural network architecture for high-performance artificial intelligence applications | |
CN112396176B (zh) | 一种硬件神经网络批归一化系统 | |
Wang et al. | Complementary Digital and Analog Resistive Switching Based on AlOₓ Monolayer Memristors for Mixed-Precision Neuromorphic Computing | |
Peng et al. | Ferroelectric-like non-volatile FET with amorphous gate insulator for supervised learning applications | |
Lin et al. | Resistive memory-based zero-shot liquid state machine for multimodal event data learning | |
Zhou et al. | Thermally stable threshold selector based on CuAg alloy for energy-efficient memory and neuromorphic computing applications | |
Tang et al. | Improved crossbar array architecture for compensating interconnection resistance: Ferroelectric HZO-based synapse case | |
US11877524B2 (en) | Nanotip filament confinement |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |