A method, system and storage medium for solving equations

Abstract

The present invention provides a method, system and storage medium for solving equations. In a coordinate system, a strength is assigned to a point on a curve of each equation in a set of equations to be solved; wherein the strength of the point gradually decays over time and can be superimposed; after a preset time, the strength of the point on the curve of each equation is obtained; the intersection of the curves is the solution to the set of equations, and since the strength of the point can be superimposed, the strength at the intersection is higher, and the strength of all points decays at the same rate, the point with the largest strength is taken as the solution to the set of equations to be solved. It can be seen that the present invention does not require iterative calculations, and utilizes the characteristic that the strength gradually decays over time. It only takes a certain amount of time to obtain the solution to the set of equations to be solved, and it has nothing to do with the complexity of the set of equations to be solved, and the power consumption is very low.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method, system and storage medium for solving equations. Background Art

[0002] At present, the iterative method is generally used by computers to solve complex problems such as systems of equations. The basic steps of the iterative method are as follows:

[0003] Given a task: Let A∈R n×n Non-singular, b∈R n , solve the system of equations Ax=b.

[0004] Step 1: Construct an equivalent system of equations with the same solution x=Gx+d through some transformation.

[0005] Step 2: Create an iteration formula: x (k+1) =Gx (k) +d, k=0,1,…

[0006] Step 3: Select the initial vector x (0) ={x 1 (0) , x 2 (0) , ..., x n (0)} T , repeatedly use the iterative formula to iterate, hoping to approach the exact solution until the accuracy requirement is met.

[0007] The more complex the equation, the more calculation steps are required. Although the computer runs very fast, it will cause a lot of power consumption, and as the order and number of equations increase, a lot of computing resources are needed. Summary of the invention

[0008] The present application provides a method, system and storage medium for solving equations, aiming to reduce power consumption of solving equations.

[0009] According to the first aspect, an embodiment provides a method for solving an equation, comprising the following steps:

[0010] In the coordinate system, an intensity is assigned to the point on the curve of each equation in the system of equations to be solved; the intensity of the point gradually decays over time and can be superimposed;

[0011] After a preset time, the intensity of the points on the curve of each equation is obtained; and the point with the largest intensity is taken as the solution of the set of equations to be solved.

[0012] In the method, the coordinate system is composed of an array of bionic synaptic devices with short-range plasticity, the bionic synaptic devices are points in the coordinate system, and the weight of the bionic synaptic devices is the strength of the point.

[0013] In the method, taking the point with the largest intensity as the solution of the set of equations to be solved comprises:

[0014] Compare the strengths of the points to obtain the point with the greatest strength; there are one or more points with the greatest strength; use the point with the greatest strength as the solution of the set of equations to be solved; or,

[0015] The intensity of each point is clustered, and the point of the largest intensity is used as the solution of the equation group to be solved.

[0016] According to a second aspect, an embodiment provides a system for solving equations, comprising:

[0017] An array of bionic synaptic devices for simulating a coordinate system; the bionic synaptic devices are points in the coordinate system;

[0018] The processing module is used to assign a weight to the bionic synaptic device corresponding to the point on the curve of each equation in the set of equations to be solved in the bionic synaptic device array; after a preset time, obtain the weight of the bionic synaptic device corresponding to the point on the curve of each equation; and output the point with the largest weight as the solution of the set of equations to be solved.

[0019] In the system, the processing module is also used to obtain the set of equations to be solved through a human-computer interaction device, and display the solution of the set of equations to be solved through the human-computer interaction device.

[0020] In the system, the processing module uses the point with the largest weight as the solution of the set of equations to be solved, including:

[0021] Compare the weights of the bionic synaptic devices to obtain the bionic synaptic device with the largest weight; there are one or more bionic synaptic devices with the largest weight; take the point corresponding to the bionic synaptic device with the largest weight as the solution of the equation group to be solved; or,

[0022] The weights of the bionic synaptic devices are clustered, and the points corresponding to the type of bionic synaptic devices with the largest weights are taken as the solutions to the set of equations to be solved.

[0023] In the system, the bionic synaptic device has short-range plasticity.

[0024] According to a third aspect, an embodiment provides a computer-readable storage medium, on which a program is stored, and the program can be executed by a processor to implement the method as described above.

[0025] According to the method, system and storage medium for solving equations in the above-mentioned embodiment, in the coordinate system, an intensity is assigned to the point on the curve of each equation in the set of equations to be solved; wherein the intensity of the point gradually decays over time and can be superimposed; after a preset time, the intensity of the point on the curve of each equation is obtained; the intersection of the curves is the solution of the set of equations, and since the intensity of the point can be superimposed, the intensity at the intersection is higher, and the intensity of all points decays at the same rate, the point with the largest intensity is taken as the solution of the set of equations to be solved. It can be seen that the present invention does not require iterative calculations, and utilizes the characteristic that the intensity gradually decays over time, and only needs to wait for a certain period of time to obtain the solution of the set of equations to be solved, and it has nothing to do with the complexity of the set of equations to be solved, and the power consumption is very low. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flow chart of an embodiment of a method for solving an equation provided by the present invention;

[0027] Figure 2 A structural block diagram of an embodiment of a system for solving equations provided by the present invention;

[0028] Figure 3 A schematic diagram of a bionic synaptic device array in the system for solving equations provided by the present invention;

[0029] Figure 4 A structural block diagram of an embodiment of a processing module in the system for solving equations provided by the present invention;

[0030] Figure 5 A schematic diagram of the point intensity assignment on the curve corresponding to the equation in the system for solving the equation provided by the present invention;

[0031] Figure 6 A schematic diagram of the system for solving equations provided by the present invention after intensity assignment for a period of time;

[0032] Figure 7 A schematic diagram of obtaining solutions to a system of equations in the system for solving equations provided by the present invention. DETAILED DESCRIPTION

[0033] The present invention will be further described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0034] The present invention is inspired by the calculation process of the human brain. The human brain can receive a large amount of information at the same time, but it will not store all the information as memory. After the stimulation of a large amount of information, only events with a certain number of stimulations can be stored in the memory, and the remaining events will gradually be forgotten. The human brain uses this process to solve problems. It receives a large number of events at the same time, and then waits for a certain period of time. The data that is not a solution (a small number of stimulations) will automatically disappear, and the solution to the problem (data with multiple stimulations) will automatically remain, that is, the solution to the problem is obtained. The power consumption of the entire human brain is only about 20W, but it can handle a large number of problems based on this calculation process. This calculation method only requires extremely low energy consumption. After receiving the stimulus, it can automatically obtain the solution to a complex problem by waiting for a certain period of time. It has obvious advantages in solving complex problems such as multi-order equations.

[0035] like Figure 1 As shown, the method for solving the equation provided by the present invention comprises the following steps:

[0036] Step 1: In the coordinate system, assign an intensity to the point on the curve of each equation in the set of equations to be solved; wherein the intensity of the point gradually decays over time and can be superimposed.

[0037] Step 2. After a preset time, obtain the intensity of the points on the curve of each equation; take the point with the largest intensity as the solution to the system of equations to be solved. The intersection of the curves is the solution to the system of equations. Since the intensity of the points can be superimposed, the intensity at the intersection is higher, and the intensity of all points decays at the same rate, so the point with the largest intensity is the solution to the system of equations to be solved. There may be one solution to the system of equations to be solved, such as the intersection of two straight lines; there may also be multiple solutions (two or more) to the system of equations to be solved, such as the intersection of a straight line and the left and right branches of a hyperbola; therefore, there are one or more points with the largest intensity. In other words, if there is one point with the largest intensity, it means that the solution to the system of equations is unique, and if there are multiple points with the largest intensity, it means that there are multiple solutions to the system of equations.

[0038] There are many ways to select the point with the greatest intensity. For example, the intensity of each point can be compared to obtain the point with the greatest intensity. The intensity of each point can also be clustered, and the point with the greatest intensity is used as the solution of the set of equations to be solved.

[0039] It can be seen that the present invention does not require iterative calculations, but utilizes the characteristic that the intensity gradually decays over time. It only takes a certain amount of time to obtain the solution to the set of equations to be solved, and it is independent of the complexity of the set of equations to be solved, and the power consumption is very low.

[0040] Specifically, the present invention also provides a system for solving equations, such as Figure 2 As shown, the system is used to solve the equation. The system includes: a processing module 10, a bionic synaptic device array 20 and a human-computer interaction device.

[0041] like Figure 3As shown, the bionic synaptic device array 20 is composed of a plurality of bionic synaptic devices 210 arrays. The bionic synaptic device array 20 is used to simulate a coordinate system. The coordinate system can be a two-dimensional coordinate system, a three-dimensional space coordinate system, etc. This embodiment is described by taking a two-dimensional rectangular coordinate system as an example. The bionic synaptic device 210 is a point in the coordinate system; the bionic synaptic device 210 can be a point or several points in the coordinate system, depending on the accuracy; or a point in the coordinate system can be formed by a plurality of bionic synaptic devices 210, and the average of the strengths (e.g., weights) of the plurality of bionic synaptic devices 210 is used as the strength of the point to improve the accuracy. The bionic synaptic device 210 can have short-range plasticity, such as a memristor, a synaptic device based on a transistor (such as an electric double-layer thin-film transistor EDLTFTs), a synaptic device based on a 2D material, a three-terminal synaptic device, etc. The short-range plasticity of the bionic synaptic device 210 means that after a certain pulse stimulation, the weight (current or other defined parameters) of the device will change, but this change will decay over time. This characteristic can fully adapt to the present calculation method and has strong practical operability.

[0042] The human-computer interaction device is used to receive user input and output visual information; it can receive user input using an input device such as a keyboard, mouse, touchpad, etc., or it can use a touch screen integrated with a display; it can output visual information using a display.

[0043] The processing module 10 is used to assign a weight to the bionic synaptic device corresponding to the point on the curve of each equation in the to-be-solved equation group in the bionic synaptic device array; after a preset time, obtain the weight of the bionic synaptic device corresponding to the point on the curve of each equation; and output the point with the largest weight as the solution of the to-be-solved equation group. The following takes a two-variable linear equation group as an example to illustrate the process of the processing module 10 controlling the system to solve the equation.

[0044] Step 1 specifically includes: the user inputs the equations to be solved through the input module 310, which is a set of linear equations of two variables in this embodiment. Figure 4 As shown, in one embodiment, the processing module 10 includes a processor 110, a write circuit 120 and a read circuit 130. The write circuit 120 is used to perform pulse stimulation on the bionic synaptic device and assign it a weight. The read circuit 130 is used to detect the weight of the bionic synaptic device. The processor 110 is used to control the system to complete the solution of the equation. The processor 110 receives the system of equations to be solved input by the user through the input module 310. Each equation of the system of equations can be represented by a curve in a rectangular coordinate system. Each bionic synaptic device 210 can be written and read through a certain voltage operation. Therefore, in the rectangular coordinate system, the processor 110 performs pulse stimulation (equivalent to writing) on ​​the bionic synaptic device 210 corresponding to the point on the curve of each equation in the system of equations to be solved through the write circuit 120, as shown Figure 5 shown.

[0045] Step 2 specifically includes: as time changes, the weights (current or other parameters) of the bionic synaptic device 210 will be slowly forgotten. Each bionic synaptic device 210 is the same, so its weight decays at the same rate. After the preset time, the processor 110 reads the weights of the bionic synaptic device 210 corresponding to each point on the curve through the reading circuit 130 with a smaller reading voltage (smaller than the writing voltage), and compares the sizes, and takes the point with the largest weight as the solution to the set of equations to be solved. The preset time can be set according to the short-range plasticity of the bionic synaptic device 210, or user needs, etc.

[0046] Taking a system of linear equations in two variables as an example, assuming that the weights of all points on a straight line represented by an equation are 1, then the weight of the intersection point is 2 because it belongs to two straight lines, such as Figure 5 As shown. Initially, the weight ratio of solution points to non-solution points is 2:1. As time changes, the weights of all points gradually decay. When the weights of all points decay by 0.5, the weight of non-solution points is 0.5, while the weight of solution points is 1.5. At this time, the weight ratio of solution points to non-solution points is 3:1, as shown in Figure 6 When the weight decays by 0.9, the weight of the non-solution point is 0.1, and the weight of the solution is 1.1. The weight ratio is 11:1. Figure 7 Therefore, as time changes, the weight ratio of the solution to the non-solution will gradually increase, and the value of the solution will be automatically obtained over time.

[0047] It can be seen that waiting for a certain period of time can better identify the solution point. There are many ways to identify the point with the largest weight. For example, in one embodiment, the processor 110 can compare the weights of each bionic synaptic device 210 to obtain the bionic synaptic device with the largest weight; there are one or more bionic synaptic devices with the largest weight; the point corresponding to the bionic synaptic device with the largest weight is used as the solution to the system of equations to be solved. In another embodiment, the processor 110 can cluster the weights of each bionic synaptic device 210 and use the point corresponding to the type of bionic synaptic device with the largest weight as the solution to the system of equations to be solved. For a system of linear equations in two variables, the coordinates (x value and y value) of the point with the largest weight is the solution to the system of equations. The processor 110 also displays the solution to the system of equations to be solved through the display 320 for user viewing.

[0048] If the existing computer calculation method is used, it takes multiple iterations to find the correct x, y combination, but with this method, you only need to perform initial stimulation and wait for a while to get the answer. This method can not only calculate simple two-variable linear equations, but also for complex multi-dimensional equations, calculation problems with multiple solutions, etc., this method can effectively save computing resources. And because this calculation method only needs to perform initial intensity assignment, the power consumption required is very small, which has great advantages in calculating complex problems.

[0049] Those skilled in the art will appreciate that all or part of the steps of the various methods in the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which may include: a read-only memory, a random access memory, a disk or an optical disk, etc.

[0050] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A method for solving an equation, characterized in that: The steps include: In a coordinate system, a point on a curve of each equation in a set of equations to be solved is given an intensity; wherein the intensity of the point gradually decays over time and can be superimposed; the coordinate system is composed of an array of bionic synaptic devices with short-range plasticity, the bionic synaptic devices are points in the coordinate system, and the weight of the bionic synaptic devices is the intensity of the point; After a preset time, the intensity of the points on the curve of each equation is obtained; and the point with the largest intensity is taken as the solution of the set of equations to be solved.

2. The method according to claim 1, characterized in that The step of taking the point with the largest intensity as the solution of the set of equations to be solved comprises: Compare the strengths of the points to obtain the point with the greatest strength; there are one or more points with the greatest strength; use the point with the greatest strength as the solution of the set of equations to be solved; or, The intensity of each point is clustered, and the point of the largest intensity is used as the solution of the equation group to be solved.

3. A system for solving equations, characterized in that include: An array of biomimetic synaptic devices for simulating coordinate systems; The bionic synaptic device is a point in a coordinate system; A processing module is used to assign a weight to the bionic synaptic device corresponding to the point on the curve of each equation in the set of equations to be solved in the bionic synaptic device array; after a preset time, obtain the weight of the bionic synaptic device corresponding to the point on the curve of each equation; take the point with the largest weight as the solution of the set of equations to be solved and output it; wherein the bionic synaptic device has short-range plasticity, and the weights assigned to the bionic synaptic device can be superimposed and gradually decay over time.

4. The system according to claim 3, characterized in that The processing module is also used to obtain the set of equations to be solved through the human-computer interaction device, and display the solution of the set of equations to be solved through the human-computer interaction device.

5. The system according to claim 3, characterized in that The processing module uses the point with the largest weight as the solution of the set of equations to be solved, including: Compare the weights of the bionic synaptic devices to obtain the bionic synaptic device with the largest weight; there are one or more bionic synaptic devices with the largest weight; take the point corresponding to the bionic synaptic device with the largest weight as the solution of the equation group to be solved; or, The weights of the bionic synaptic devices are clustered, and the points corresponding to the type of bionic synaptic devices with the largest weights are taken as the solutions to the set of equations to be solved.

Images