Associative Learning with Temporal Contiguity in a Memristive Circuit for Large‐Scale Neuromorphic Networks

Li, Yang; Xu, Lei; Zhong, Yingpeng; Zhou, Ya-Xiong; Zhong, Shujing; Hu, Yang-Zhi; Chua, Leon O.; Miao, Xiangshui

doi:10.1002/aelm.201500125

Cited by 75 publications

(57 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the past few years, ionotronic transistors or memristors were proposed to mimic advanced neural functions. By connecting pressure sensors and oscilloscope with neuromorphic devices and adoping mathematical modeling and software simulation, advanced neural functions are simulated, including logic function, image memorization, pattern recognition, face recognition, classic conditioning, tactile‐perception system, etc. In this section, we shortly discuss these achievements.…”

Section: Advanced Neural Functions Based On Ionotronic Neuromorphic Dmentioning

confidence: 99%

Ionotronic Neuromorphic Devices for Bionic Neural Network Applications

Zhu

2019

Physica Rapid Research Ltrs

View full text Add to dashboard Cite

The von Neumann bottleneck constrains developments of conventional artificial intelligences (AIs) toward portable, energy efficient, and truly brain‐like systems. Biological synapses connecting neurons deliver neural action potentials by regulating neurotransmitters between presynaptic and postsynaptic membranes. In a similar way, ionotronic transistors and memristors transmit electronic signals through the migration of ionic species in dielectric and resistive switching electrolyte under external electrical field. Thus, utilizing ionotronic devices to emulate synapses and building bionic neural networks opens up a brand‐new path to realize hardware‐based AI. Moreover, it would allow the fabrication of an artificial perception learning system to mimic the human perception system with multi‐sensory learning abilities. This review presents ionotronic devices for neuromorphic engineering applications. The operation mechanisms and brain‐inspired synaptic responses and neural functions are discussed. At last, outlooks for ionotronic devices to construct bionic neural networks are provided.

show abstract

Section: Advanced Neural Functions Based On Ionotronic Neuromorphic Dmentioning

confidence: 99%

Ionotronic Neuromorphic Devices for Bionic Neural Network Applications

Zhu

2019

Physica Rapid Research Ltrs

View full text Add to dashboard Cite

show abstract

“…In addition to the application in logic‐in‐memory computing, the XOR and XNOR logic opertations also play a significant role in pattern recognition systems . XOR and XNOR are not linearly separable with multiple decision boundaries, which makes XOR and XNOR logic operations regard as matching.…”

Section: Resultsmentioning

confidence: 99%

“…However, the ratio of HRS/LRS and device retention is less important for logic application. High reliability of switching property and controllable gradual change can also be considered for accurate neural applications .…”

Section: Resultsmentioning

confidence: 99%

“…These Ag 150/AgInSbTe 20/Ta 200 (nm) crossbar memristors were fabricated by ultraviolet photolithography, magnetron sputtering, and lift‐off processes. Fabrication details can be found in our previous publications . Figure (b) shows the surface morphology of the electrolyte film AgInSbTe measured by atomic force microscopy (AFM, Shimadzu SPM‐9700), indicating small surface roughness with root mean square (RMS) of 0.414 nm.…”

Section: Memristor Specification and Logic Operation Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A hybrid memristor-CMOS XOR gate for nonvolatile logic computation

Zhou

et al. 2015

Phys. Status Solidi A

Self Cite

View full text Add to dashboard Cite

Memristor-based logic gates that can execute memory and logic operations are promising elements for building non-von Neumann computation architecture. In this paper, we proposed a fundamental hybrid memristor-CMOS nonvolatile XOR Boolean logic block with one memristor and four voltagecontrolled switches, such as MOSFET or transmission gate. The Ag/Ag 5 In 5 Sb 60 Te 30 /Ta structure memristors with bipolar resistive switching behaviors were utilized to experimentally demonstrate the XOR function of the block. Further, with the XOR logic block as a basis, a full adder as an extensive circuit example was designed and verified by HSPICE simulation. Our work shows logic-in-memory capabilities with memristors, and provides an alternative way to implement nonvolatile logic operations and construct parallel computation systems.

show abstract

“…In neural networks, association corresponds to an increase of synaptic strength, which is controlled by action potentials emitted by pre-and postsynaptic neurons. 2,3 Associative learning can be emulated with memristive Hopfield neural networks, 4 memristive circuits 5,6,7,8 or magnetic tunnel junctions. 9 Other demonstrations of associative learning implement two artificial synapses with memristor emulators 10 or a resistor and a memristor.…”

mentioning

confidence: 99%

Associative learning with Y-shaped floating gate transistors operated in memristive modes

Maier

Hartmann

Emmerling

et al. 2017

Applied Physics Letters

View full text Add to dashboard Cite

2Associative learning in neural networks is crucial and required to relate new experience with existing memories and was first observed by I. Pavlov in his famous dog experiment. 1 Stimulating different senses by the sight of food and the sound of a bell, I. Pavlov trained his dog to correlate salivation with the sound of the bell (association), although at the beginning of the experiment salivation was only triggered by the sight of food. In neural networks, association corresponds to an increase of synaptic strength, which is controlled by action potentials emitted by pre-and postsynaptic neurons. 2,3 Associative learning can be emulated with memristive Hopfield neural networks, 4 memristive circuits 5,6,7,8 or magnetic tunnel junctions. 9 Other demonstrations of associative learning implement two artificial synapses with memristor emulators 10 or a resistor and a memristor. 11 Artificial synapses based on memristors enable synaptic modifications by applying pre-and postsynaptic voltage pulses to the two terminals of the device, 12,13 which allows to combine information storage and processing on the same physical platform. 14,15,16 Owing to the two-terminal character, however, the application of postsynaptic voltage pulses blocks the device output and prevents information to be transferred during the learning process. The devices separate signal transmission and learning. Hence,18,19 or even fourterminal 20 synaptic devices may be beneficial because of their ability to decouple the postsynaptic pulse from the output. Indeed, associative learning with simultaneous learning and signal transmission has been demonstrated with three-terminal synapses based on nano-particle organic memory field effect transistors by applying postsynaptic pulses with amplitudes as large as 30 V to the third terminal. 21 In memristor-synapses, extra terminals may be implemented with additional gates that allow tuning the set voltage with an electric field effect. 22 We report associative learning with Y-shaped quantum dot floating gate transistors operated in memristive modes. The geometry of the devices with two input terminals and one output terminal enables realizing associative learning with postsynaptic voltage amplitudes as low as 0.5 V. The 3 conductance of the Y-shaped quantum dot floating gate transistor depends on the amount of localized charges on quantum dots (QDs) that are precisely positioned in the two input terminals. Emulating external stimuli "food" and "bell" with two input voltages allows implementing an association between "bell" and "salivation". The number of required pulses to develop (association) or to forget (extinction) the correlation between "bell" and "salivation" is controlled by the amplitudes of the input voltages. Connecting the voltage applied to the drain contact (i.e. right branch V r ) with the side gates (see Fig. 1(a)) leads to a memristive operation. 25,26,27 The Y-shaped floating gate transistor is characterized by applying the voltage V r = V l to both branches with resistances of R l = R...

show abstract

Associative Learning with Temporal Contiguity in a Memristive Circuit for Large‐Scale Neuromorphic Networks

Cited by 75 publications

References 45 publications

Ionotronic Neuromorphic Devices for Bionic Neural Network Applications

Ionotronic Neuromorphic Devices for Bionic Neural Network Applications

A hybrid memristor-CMOS XOR gate for nonvolatile logic computation

Associative learning with Y-shaped floating gate transistors operated in memristive modes

Contact Info

Product

Resources

About