Probability flipping-based polymorphic neuronal synaptic device and modulation method thereof

Abstract

The invention provides a polymorphic synaptic device based on probability flipping and a modulation method thereof, and relates to the technical field of semiconductor devices. The polymorphic neurosynaptic device comprises a plurality of probability devices which are connected in series; wherein the resistance state of each probabilistic device presents a probabilistic high resistance state or low resistance state according to different input pulse current densities; and simulating the weight of the neuronal synapse by using the total voltage of series connection of a plurality of probability devices or the total output voltage of multiple pulse currents of the same probability device, thereby realizing the artificial synapse device for neuromorphic calculation. The polymorphic neuronal synaptic device can enable neuromorphic calculation to be transformed from deterministic simulation to random bionic simulation, and has unique advantages in scenes such as adaptive learning and uncertain reasoning.

Description

Technical Field

[0001] The present invention relates to the technical field of semiconductor devices, and in particular to a polymorphic neural synaptic device based on probability flipping and a modulation method thereof. Background Art

[0002] In recent years, neuromorphic computing technology, which mimics the human brain, has received widespread attention. In order for semiconductor devices to implement brain-like information transmission methods, a high-performance artificial synaptic device that can simulate various synaptic connections is needed.

[0003] In a probabilistic device, when subjected to an out-of-plane magnetic field and injected with a varying pulsed current, a deterministic magnetization reversal occurs. Reducing thermal stability can cause the magnetic layer to undergo random magnetization reversals when the current is injected. This coexistence of deterministic and probabilistic magnetization reversals, driven by the spin-orbit torque (SOT) or spin-transfer torque (STT) effects, is known as the rebound phenomenon.

[0004] By utilizing the characteristics of the above-mentioned probabilistic devices, artificial synaptic devices for neuromorphic computing can be realized. Summary of the Invention

[0005] The purpose of the present invention is to provide a polymorphic neural synaptic device based on probability flipping and its modulation method, which utilizes the multi-resistance state of the total voltage of multiple probabilistic devices in series or the total output voltage of multiple pulse currents of the same probabilistic device to simulate the weight of neuronal synapses and realize artificial synaptic devices for neuromorphic computing.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] On the one hand, a polymorphic neural synapse device based on probability flipping is provided, which includes multiple probabilistic devices connected in series; wherein the resistance state of each probabilistic device presents a probabilistic high-resistance state or a low-resistance state according to the different input pulse current densities; the weight of the neuronal synapse is simulated by the total voltage of the series connection of multiple probabilistic devices or the total output voltage of multiple pulse currents of the same probabilistic device.

[0008] Optionally, the probabilistic device is a magnetic tunnel junction device or a Hall bar device.

[0009] Optionally, the magnetic tunnel junction device includes, from bottom to top, a substrate, a spin current generating layer, a ferromagnetic layer, an intermediate non-magnetic insulating barrier layer, a magnetic pinning layer, and a readout electrode layer.

[0010] Optionally, the Hall bar device includes, from bottom to top, a substrate, a first spin current generating layer, a second spin current generating layer, a ferromagnetic layer, and a covering layer.

[0011] Optionally, the probabilistic device undergoes different magnetization reversals for different pulse current densities:

[0012] When the input pulse current density is greater than the critical reversal current density and less than the first threshold current density J0, a deterministic reversal occurs; when the input pulse current density is greater than the first threshold current density J0, a probabilistic reversal occurs; when the input pulse current density continues to increase and exceeds the second threshold current density J1, a random reversal occurs, that is, the probability of the magnetization state flipping up and down is 50%.

[0013] Optionally, when a deterministic flip occurs, the input pulse current density is 10×10 10 A.m -2 ~20×10 10 A.m -2 When the probability reversal occurs, the input pulse current density is 20×10 10 A.m -2 ~30×10 10 A.m -2 When random flip occurs, the input pulse current density is greater than 30×10 10 A.m -2 , and is less than the maximum current density that the probabilistic device can withstand.

[0014] In another aspect, a method for modulating a polymorphic neural synapse device according to any one of the above is provided, the method comprising:

[0015] For N probabilistic devices, let high resistance state be 1 and low resistance state be 0;

[0016] Input pulse current density J to each of the probabilistic devices m The pulse current density corresponds to the flip probability p m , then the total output voltage is expected to be N×p m , realize multi-resistance state regulation;

[0017] Alternatively, different pulse current densities J are input to each of the probabilistic devices. n The pulse current corresponds to different flip probability p n , realizing multi-resistance state regulation.

[0018] Optionally, a probabilistic device is reused to replace multiple probabilistic devices; the pulse current density input each time is the same or different, and the number of resistance states depends on the number of input pulse currents, so as to achieve multi-resistance state regulation.

[0019] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0020] The present invention provides a polymorphic neural synaptic device based on probability flipping and its modulation method. The polymorphic neural synaptic device is composed of multiple probabilistic devices connected in series, and the resistance state of each probabilistic device presents a probabilistic high-resistance state or a low-resistance state according to the different input pulse current densities. By utilizing the multi-resistance state of the total voltage of the series connection of multiple probabilistic devices or the total output voltage of multiple pulse currents of the same probabilistic device, the weight of the neuronal synapse can be simulated, thereby realizing an artificial synaptic device for neuromorphic computing. The polymorphic neural synaptic device can enable the paradigm shift of neuromorphic computing from deterministic simulation to stochastic biomimetic simulation, and has unique advantages in scenarios such as adaptive learning and uncertain reasoning. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 1 is a schematic structural diagram of a polymorphic neural synapse device provided by an embodiment of the present invention;

[0023] Figure 2a-2b Schematic diagram of a magnetic tunnel junction device based on SOT and STT provided in an embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of a Hall bar device provided by an embodiment of the present invention;

[0025] Figure 4 is a schematic diagram of the rebound phenomenon provided by an embodiment of the present invention;

[0026] Figure 5 Schematic diagram of a probability curve varying with current density provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] The embodiment of the present invention provides a polymorphic neural synaptic device based on probability flipping, such as Figure 1As shown, the polymorphic neural synaptic device comprises multiple probabilistic devices connected in series, with the "+" sign indicating a series connection. The resistance state of each probabilistic device exhibits a probabilistic high-resistance or low-resistance state depending on the input pulse current density. The total voltage of the series-connected probabilistic devices, or the total output voltage of multiple pulse currents from the same probabilistic device, is used to simulate the weight of the neuronal synapse, thereby realizing an artificial synaptic device for neuromorphic computing.

[0029] In an embodiment of the present invention, the probabilistic device is a magnetic tunnel junction device (MTJ) or a Hall bar device.

[0030] like Figure 2a-2b The following are schematic diagrams of magnetic tunnel junction devices based on the spin-orbit torque effect (SOT) and the spin transfer torque effect (STT). The magnetic tunnel junction device comprises, from bottom to top, a substrate 1, a spin current generation layer 2, a ferromagnetic layer 3, an intermediate non-magnetic insulating barrier layer 4, a magnetic pinning layer 5, and a readout electrode layer 6. The arrows in the figure represent the first current (write current I in ) and the second current (readout current V out ) direction. Figure 2a-2b The spin current generation layer 2 and the ferromagnetic layer 3 are sputtered separately to reduce thermal stability. At this time, when a pulse current is injected, based on the spin-orbit torque effect (SOT) or the spin transfer torque effect (STT), deterministic reversal and probabilistic magnetization reversal will coexist, which is the rebound phenomenon.

[0031] Figure 3 Schematic diagram of a Hall bar device, which includes, from bottom to top, a substrate 1, a first spin current generating layer 7, a second spin current generating layer 8, a ferromagnetic layer 3, and a cover layer 9. Similarly, the arrows represent the first current (write current I in ) and the second current (readout current V out ) direction. Figure 3 The first spin current generating layer 7 and the second spin current generating layer 8 are sputtered separately to reduce thermal stability. When a pulse current is injected at this time, a phenomenon of deterministic reversal and probabilistic magnetization reversal will occur at the same time, which is the rebound phenomenon.

[0032] The present invention designs a polymorphic neural synaptic device that can be used for neuromorphic computing based on the rebound phenomenon of probabilistic devices. Figure 4 This is a schematic diagram of the rebound phenomenon. The probabilistic device undergoes different magnetization reversals for different pulse current densities:

[0033] When the input pulse current density is greater than the critical reversal current density and less than the first threshold current density J0, a deterministic reversal occurs; when the input pulse current density is greater than the first threshold current density J0, a probabilistic reversal occurs; when the input pulse current density continues to increase and exceeds the second threshold current density J1, a random reversal occurs, that is, the probability of the magnetization state flipping up and down is 50%. Figure 4 In the equation, ±J0 is a deterministic flip, ±J0 to ±J1 is a probabilistic flip, and anything greater than ±J1 is a random flip. The probability curve that changes with current density is as follows: Figure 5 shown.

[0034] As an optional implementation, when a deterministic flip occurs, the input pulse current density is 10×10 10 A.m -2 ~20×10 10 A.m -2 (Depending on the size of the device, the range of deterministic flipping will vary); when probabilistic flipping occurs, the input pulse current density is 20×10 10 A.m -2 ~30×10 10 A.m -2 (Depending on the device size, the range of probability flipping will vary and will not coincide with the deterministic flipping current density); when random flipping occurs, the input pulse current density is greater than 30×10 10 A.m -2 , and is less than the maximum current density that the probabilistic device can withstand, that is, the upper limit of the current density depends on the current density that the probabilistic device can withstand.

[0035] Based on the above characteristics, in an embodiment of the present invention, the multi-resistance state of the total voltage of multiple probabilistic devices in series or the total output voltage of multiple pulse currents of the same probabilistic device is used to simulate the weight of neuronal synapses and construct a polymorphic neural synaptic device based on probability flipping.

[0036] Furthermore, the present invention also provides a method for modulating the polymorphic neural synapse device, comprising:

[0037] For N probabilistic devices, let high resistance state be 1 and low resistance state be 0;

[0038] Input pulse current density J to each of the probabilistic devices m The pulse current density corresponds to the flip probability p m , then the total output voltage is expected to be N×p m , realize multi-resistance state regulation;

[0039] Alternatively, different pulse current densities J are input to each of the probabilistic devices. nThe pulse current corresponds to different flip probability p n , realizing multi-resistance state regulation.

[0040] As another embodiment of the present invention, a probabilistic device can be reused to replace multiple probabilistic devices; the pulse current density input each time is the same or different, and the number of resistance states depends on the number of input pulse currents to achieve multi-resistance state regulation.

[0041] Specifically, the embodiment of the present invention provides the following modulation scheme:

[0042] Solution 1: For N probability devices, give each probability device J m The pulse current density corresponds to the probability p m , then the total output voltage is expected to be N×p m Theoretically, if there are enough probabilistic devices and the voltage values ​​are distinguishable, there can be countless resistance states (in traditional devices, N devices have at most N resistance states).

[0043] Solution 2: For N probabilistic devices, give each probabilistic device a different current density J n , corresponding to different probabilities p n , there can also be countless resistance states, but the adjustability of the synapse is greatly increased.

[0044] Option 3: Based on Option 1 and Option 2, multiple probabilistic devices can be replaced with one probabilistic device for repeated use. The current density injected each time can be the same or different, which increases the operability of neuromorphic computing. The number of resistance states depends on the number of input currents, and theoretically there is no upper limit to the number of resistance states.

[0045] Compared with the prior art, the present invention provides a new polymorphic neural synaptic device based on probability flipping, which is composed of multiple probabilistic devices connected in series. The resistance state of each probabilistic device presents a probabilistic high-resistance state or a low-resistance state according to the different input pulse current densities. By utilizing the multi-resistance state of the total voltage of multiple probabilistic devices in series or the total output voltage of multiple pulse currents of the same probabilistic device, the weights of neuronal synapses can be simulated, thereby realizing an artificial synaptic device for neuromorphic computing. The polymorphic neural synaptic device can enable neuromorphic computing to shift from a paradigm of deterministic simulation to a paradigm of stochastic biomimetic simulation, and has unique advantages in scenarios such as adaptive learning and uncertain reasoning. It may become a key component of the next generation of brain-like chips.

[0046] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the element.

[0047] References in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include a particular feature, structure, or characteristic, but not every embodiment necessarily includes that particular feature, structure, or characteristic. In addition, when a particular feature, structure, or characteristic is described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement that feature, structure, or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0048] It should be understood that the term "and / or" as used herein simply describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the related objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0049] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0050] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0051] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0052] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0053] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0054] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0055] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0056] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A polymorphic neural synaptic device based on probability flipping, characterized in that: The polymorphic neural synapse device includes multiple probabilistic devices connected in series; wherein, the resistance state of each probabilistic device presents a probabilistic high-resistance state or a low-resistance state according to the different input pulse current densities; the weight of the neuronal synapse is simulated by the total voltage of the series connection of multiple probabilistic devices or the total output voltage of multiple pulse currents of the same probabilistic device.

2. The polymorphic neural synapse device according to claim 1, wherein: The probabilistic device is a magnetic tunnel junction device or a Hall bar device.

3. The polymorphic neural synapse device according to claim 2, wherein: The magnetic tunnel junction device comprises, from bottom to top, a substrate, a spin current generating layer, a ferromagnetic layer, an intermediate non-magnetic insulating barrier layer, a magnetic pinning layer and a readout electrode layer.

4. The polymorphic neural synapse device according to claim 2, wherein: The Hall bar device comprises, from bottom to top, a substrate, a first spin current generating layer, a second spin current generating layer, a ferromagnetic layer and a covering layer.

5. The polymorphic neural synapse device according to claim 1, wherein: The probabilistic device undergoes different magnetization reversals for different pulse current densities: When the input pulse current density is greater than the critical reversal current density and less than the first threshold current density J0, a deterministic reversal occurs; when the input pulse current density is greater than the first threshold current density J0, a probabilistic reversal occurs; when the input pulse current density continues to increase and exceeds the second threshold current density J1, a random reversal occurs, that is, the probability of the magnetization state flipping up and down is 50%.

6. The polymorphic neural synapse device according to claim 5, characterized in that: When a deterministic flip occurs, the input pulse current density is 10×10 10 A.m -2 ~20×10 10 A.m -2 When the probability reversal occurs, the input pulse current density is 20×10 10 A.m -2 ~30×10 10 A.m -2 When random flip occurs, the input pulse current density is greater than 30×10 10 A.m -2 , and is less than the maximum current density that the probabilistic device can withstand.

7. A modulation method based on the polymorphic neural synapse device according to any one of claims 1 to 6, characterized in that: The method comprises: For N probabilistic devices, let high resistance state be 1 and low resistance state be 0; Input pulse current density J to each of the probabilistic devices m The pulse current density corresponds to the flip probability p m , then the total output voltage is expected to be N×p m , realize multi-resistance state regulation; Alternatively, different pulse current densities J are input to each of the probabilistic devices. n The pulse current corresponds to different flip probability p n , realizing multi-resistance state regulation.

8. The method for modulating a polymorphic neural synapse device according to claim 7, wherein: A probabilistic device is reused to replace multiple probabilistic devices; the pulse current density input each time is the same or different, and the number of resistance states depends on the number of input pulse currents, so as to realize multi-resistance state regulation.

Images