Error training scheme determination method and device, equipment and medium

Abstract

The invention discloses an error training scheme determination method and device, equipment and a medium. The method comprises the following steps: acquiring target training reference information of a to-be-trained teller; the training reference information comprises personal information of a teller and an error record of the teller on counter operation; based on the target training reference information, determining a target training scheme of the error training through an intelligent judgment model; the target training scheme comprises a training theme of the error training and to-be-trained tellers needing to participate in the error training. According to the embodiment of the invention, the training effect of error training can be improved.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a method, device, equipment and medium for determining an error training program. Background Art

[0002] When bank tellers are handling business for customers, they may make various operational errors due to carelessness, unfamiliarity with the process, and other reasons. For example, a customer deposits money but accidentally enters it as "withdrawal" in the system. We call this type of operational error an operational error. For bank tellers who frequently make operational errors, the bank will regularly organize offline centralized training courses on weekends to reduce the error rate of employees through professional guidance. The training content covers multiple areas, including but not limited to mentality adjustment, attention improvement, interpretation of bank rules and regulations, improvement of operational skills, and explanation of the principles of counter information systems. However, due to objective conditions such as training time and training teachers, each training can only focus on one topic to ensure maximum training effect.

[0003] In addition, given the individual differences among employees, their responses to and effects on training courses also vary significantly. Some employees can effectively reduce their operational error rate by participating in training; while other employees can actively reduce errors through self-adjustment even if they do not participate in training. However, some employees may feel more fatigued due to attending additional training courses on weekends, which may increase their risk of making mistakes at work. Therefore, when formulating training plans, it is necessary to fully consider the individual differences and actual needs of employees in order to achieve the optimal allocation of training resources and the best results.

[0004] Due to limited time and human resources, banks can only organize an "error prevention training class" once a month. During this period, the tellers who make mistakes and the types of mistakes they make are varied. Therefore, how to accurately determine whether employees need training and what topics to choose for training can more effectively reduce the overall error rate is not a decision that can be easily made through simple statistical analysis. Summary of the invention

[0005] The present invention provides a method, device, equipment and medium for determining an error training program to improve the training effect of error training.

[0006] According to one aspect of the present invention, a method for determining an error training scheme is provided, comprising:

[0007] Obtaining target training reference information of the teller to be trained; the training reference information includes the teller's personal information and the teller's error record in counter operations;

[0008] Based on the target training reference information, a target training program for this error training is determined through an intelligent judgment model; the target training program includes the training topic of this error training and the tellers to be trained who need to participate in this error training.

[0009] According to another aspect of the present invention, there is provided a device for determining an error training scheme, comprising:

[0010] An acquisition module is used to acquire target training reference information of the teller to be trained; the training reference information includes the teller's personal information and the teller's error record in counter operations;

[0011] A determination module is used to determine the target training program of this error training based on the target training reference information through an intelligent judgment model; the target training program includes the training topic of this error training and the tellers to be trained who need to participate in this error training.

[0012] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method for determining an error training scheme according to any one of the embodiments of the present invention is implemented.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the error training scheme described in any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining an error training scheme according to any embodiment of the present invention when executed.

[0015] The embodiment of the present invention combines big data and artificial intelligence technology, adopts a deep reinforcement learning algorithm, and combines the bank's teller data to build an intelligent judgment model. The model automatically and accurately determines the theme and candidates for each training session, greatly reducing manual dependence and improving training efficiency. It enables HR and training departments to save a lot of time and resources and focus on the innovation and quality improvement of course content. Moreover, as employee status information changes, skills improve, and the transaction environment changes, unlike simple statistical analysis, this solution can adjust the list of participating employees and course theme content in real time to ensure that the training content is always synchronized with the actual needs of employees and the direction of business development.

[0016] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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.

[0018] Figure 1 is a flow chart of a method for determining an error training program according to an embodiment of the present invention;

[0019] Figure 2A is a flow chart of a method for determining an error training program according to another embodiment of the present invention;

[0020] Figure 2B is a schematic diagram of a reinforcement learning principle provided according to another embodiment of the present invention;

[0021] Figure 2C is a schematic diagram of a proximal strategy optimization algorithm principle provided according to another embodiment of the present invention;

[0022] Figure 3 is a structural schematic diagram of a device for determining an error training scheme according to another embodiment of the present invention;

[0023] Figure 4 It is a schematic diagram of the structure of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] Figure 1 This is a flowchart of a method for determining an error training program provided by an embodiment of the present invention. This embodiment can be applied to the situation where the training topic of this training and which employees who have made errors recently need to participate in this training need to be determined before the date of error training. This method can be executed by a device for determining an error training program, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device with corresponding data processing capabilities. Figure 1 As shown, the method includes:

[0027] S110. Obtain target training reference information for the teller to be trained.

[0028] S120: Based on the target training reference information, determine the target training plan for this error training through an intelligent judgment model.

[0029] The training reference information includes the teller's personal information and the teller's error record in counter operation. The target training plan includes the training topic of this error training and the tellers to be trained who need to participate in this error training. The tellers to be trained are usually tellers who have recently made counter operation errors or frequently make counter operation errors. The number can be one or more.

[0030] Specifically, tellers to be trained are selected from all employees according to the screening rules. The personal information of the tellers to be trained and the error records of the tellers in the counter operations are collected from different systems as reference information for target training. The target training reference information is input into the intelligent judgment model obtained in advance through reinforcement learning. The intelligent judgment model outputs the training topic of this training and the tellers to be trained need to participate in this training. Based on this, the system obtains the contact information of the tellers to be trained and sends them the training topic and date of this training.

[0031] The embodiment of the present invention combines big data and artificial intelligence technology, adopts a deep reinforcement learning algorithm, and combines the bank's teller data to build an intelligent judgment model. The model automatically and accurately determines the theme and candidates for each training session, greatly reducing manual dependence and improving training efficiency. It enables HR and training departments to save a lot of time and resources and focus on the innovation and quality improvement of course content. Moreover, as employee status information changes, skills improve, and the transaction environment changes, unlike simple statistical analysis, this solution can adjust the list of participating employees and course theme content in real time to ensure that the training content is always synchronized with the actual needs of employees and the direction of business development.

[0032] Figure 2A This is a flowchart of a method for determining an error training program provided by another embodiment of the present invention. This embodiment is optimized and improved on the basis of the above embodiment. Figure 2A As shown, the method includes:

[0033] S210. Obtain error records of the teller to be trained in counter operations based on the counter system and operation management process.

[0034] S220. Obtain personal information of the teller to be trained based on the human resources management system.

[0035] Specifically, by using the existing banking counter system and the bank's operational management process, it is possible to easily obtain the error records made by each teller in the historical process. The record format is shown in Table 1:

[0036] Table 1

[0037] Personnel Number Wrong date Error transaction code 000887947 2023-10-01 2723 000887947 2024-05-01 7646 000837946 2024-07-01 7654 000837946 2024-08-08 2423

[0038] Through the bank's human resources management system, it is easy to obtain detailed personal information about the teller, such as gender, age, specific counter position, years of employment, marital status, and fertility status. These data can be summarized in Table 2:

[0039] Table 2

[0040]

[0041] By establishing communication with other systems and processes of the bank, training reference information can be collected automatically and efficiently, eliminating the need for manual information collection, further improving the efficiency of formulating training programs.

[0042] Based on the above embodiment, optionally, the error record includes an error code, the number of errors and the proportion of error transactions.

[0043] Specifically, error codes correspond to difference types one by one, and each error code corresponds to one error type. The number of errors is the total number of various errors made by the teller in the recent period (for example, the last month). The percentage of error transactions is the ratio of the number of errors in the recent period to the total number of transactions processed. Error codes, number of errors, and percentage of error transactions accurately represent the areas in which tellers are prone to make mistakes and the frequency of errors. The collection and analysis of these three will help the subsequent intelligent judgment model to accurately portray the teller's portrait and formulate effective training plans based on this.

[0044] S230: Based on the target training reference information, determine the target training plan for this error training through an intelligent judgment model.

[0045] Based on the above embodiment, optionally, the intelligent judgment model is a trained action strategy network.

[0046] Specifically, reinforcement learning is a branch of machine learning, which is widely used in the decision-making field due to its unique design ideas. Figure 2B As shown, in reinforcement learning, the agent interacts with the environment, makes different actions according to the state of the environment, and distinguishes the pros and cons of actions according to the rewards fed back by the environment, increases the execution of good actions, and reduces the execution of poorly executed actions, so as to continuously interact with the environment, continuously make decisions, and finally achieve the effect of optimizing the target. In the present invention, the agent is an intelligent judgment model, which determines the execution training program. The environment counts the errors of the counter staff after participating in the training and the personal information of the counter staff, evaluates the action according to the subsequent errors, gives corresponding rewards and updates the status information.

[0047] In the reinforcement learning algorithm proposed in the present invention, the state space S is defined as the training reference information of all tellers, which can be expressed as:

[0048] S={s1,s2,…,s n}

[0049] Among them, s i is a one-dimensional vector, representing the status of employee i, mainly including the employee's personal information (gender (gen), age (age), specific counter position (job_title), years of employment (work_year), marital status (marr), number of children (child_num), etc.) and teller error information (error code (e_code), number of errors (e_num), error transaction ratio (e_r), etc.), and its formal definition is as follows:

[0050] s i={id, gen, age, job_title, work_year, marr, child_num, e_code, e_num, e_r}

[0051] The action space A is defined as whether each employee participates in the training and the training topic, which can be expressed as:

[0052] A={a1,a2,…,a n ,course}

[0053] Among them, a i Indicates whether employee i participates in this training, and its value is: Course is the training topic number of this training.

[0054] The present invention adopts the Proximal Policy Optimization (PPO) algorithm as a deep reinforcement learning algorithm to perform reinforcement learning of the intelligent judgment model. PPO is an algorithm based on policy selection action, and its basic structure is the actor-critic structure. Figure 2C As shown:

[0055] The Actor network is an action policy network that receives the observed state s t As input, it outputs the action taken by the agent. The Critic network is an action evaluation network that receives the observed state s t and the reward r of environmental feedback t+1 As input, the output is the action evaluation index, which is used to guide the update of the actor network.

[0056] The loss function of the critic network in the PPO algorithm is expressed as:

[0057]

[0058] in is the advantage function, which indicates the difference between taking the action and the average value. The goal of the critic network is to minimize the advantage function, and its parameter φ is updated by gradient descent, so that the critic network's guidance to the actor network is more and more accurate. The goal of the actor network is to maximize the objective function, which is expressed as:

[0059]

[0060] in:

[0061]

[0062] π represents the strategy, and θ is the parameter that determines the strategy.θold (a t |s t ) and π θ (a t |s t ) is that its network parameters are composed of π θ (a t |s t ) is copied at intervals and is not updated. The structure is the same as π θ (a t |s t ) is the same as the clip function. t (θ) is limited to the range of 1-∈ and 1+∈, and the clip function can ensure that the gap between the new and old strategies is small.

[0063] Based on the above embodiment, optionally, it is characterized in that the training process of the action strategy network is as follows:

[0064] Input the test training reference information of the test teller into the action strategy network to obtain the test training plan for test error training;

[0065] Executing the test training program, and calculating the function value of the reward function according to the error records of the test teller thereafter;

[0066] Inputting the function value and the experimental training reference information into the action evaluation network to obtain the action evaluation index;

[0067] The action strategy network is subjected to intensive training according to the action evaluation index.

[0068] Specifically, during the training phase, some tellers who often make mistakes are selected as test tellers. The test training reference information (S t ) Input the action strategy network to obtain the experimental training plan for this experimental error training (a t ). Execute the test training program and train some or all of the test tellers. After the training is completed, collect the error records of the test tellers thereafter, analyze and observe the error records, and calculate the function value of the reward function (r t+1 ), the function value indicates whether the effect of this training is obvious. The function value and the experimental training reference information are input into the action evaluation network to obtain the action evaluation index (A t ). Finally, the action strategy network is reinforced trained according to the action evaluation index. The above process will be repeated many times until the effect of the training plan output by the action strategy network meets the expectations, or the number of training times reaches the upper limit. The trained action strategy network can be used as an intelligent judgment model later. Through the training method of reinforcement learning, the prediction accuracy of the intelligent judgment model for the training plan is improved.

[0069] Based on the above embodiment, optionally, the function value of the reward function is calculated according to the error record of the test teller thereafter, including:

[0070] Obtain the total number of transactions and total number of errors of the test teller thereafter;

[0071] The function value of the reward function is calculated according to the ratio of the total number of errors to the total number of transactions.

[0072] Specifically, the number of tellers participating in the training is usually large. If the analysis of whether the training effect is obvious focuses on individual tellers, the calculation is large and the analysis is difficult. Since the purpose of the present invention is to reduce the overall error rate of employees, when analyzing the training effect, it does not focus on individuals, but regards all the test tellers participating in the trial as a whole, determines the total number of transactions and the total number of errors of all the test tellers after that; and calculates the function value of the reward function based on the ratio of the two. The specific formula is as follows:

[0073]

[0074] Among them, the reward function is set as the negative of the overall transaction error rate, R is the function value, α is the adjustment coefficient, the purpose is to adjust the reward value so that the algorithm converges as quickly as possible and try to avoid falling into the local optimal solution, e_code_num is the total number of errors, and total_num is the total number of transactions. By analyzing the overall error rate of employees instead of focusing on individual tellers, the convergence efficiency of the algorithm during training is improved.

[0075] The embodiment of the present invention can automatically and efficiently collect training reference information by opening up communication with other systems and processes of the bank, eliminating the need for manual information collection, thereby further improving the efficiency of formulating training programs.

[0076] Figure 3 FIG. 1 is a schematic diagram of a structure of a device for determining an error training scheme provided by another embodiment of the present invention. Figure 3 As shown, the device comprises:

[0077] An acquisition module 310 is used to acquire target training reference information of the teller to be trained; the training reference information includes the teller's personal information and the teller's error record in counter operations;

[0078] The determination module 320 is used to determine the target training program of this error training through an intelligent judgment model based on the target training reference information; the target training program includes the training topic of this error training and the tellers to be trained who need to participate in this error training.

[0079] The apparatus for determining an error training program provided in the embodiment of the present invention can execute the method for determining an error training program provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0080] Optionally, the intelligent judgment model is a trained action strategy network.

[0081] Optionally, the training process of the action strategy network is as follows:

[0082] Input the test training reference information of the test teller into the action strategy network to obtain the test training plan for test error training;

[0083] Executing the test training program, and calculating the function value of the reward function according to the error records of the test teller thereafter;

[0084] Inputting the function value and the experimental training reference information into the action evaluation network to obtain the action evaluation index;

[0085] The action strategy network is subjected to intensive training according to the action evaluation index.

[0086] Optionally, the function value of the reward function is calculated according to the error record of the test teller thereafter, including:

[0087] Obtain the total number of transactions and total number of errors of the test teller thereafter;

[0088] The function value of the reward function is calculated according to the ratio of the total number of errors to the total number of transactions.

[0089] Optionally, the acquisition module includes:

[0090] The first acquisition unit is used to acquire error records of the teller to be trained in counter operations based on the counter system and operation management process;

[0091] The second acquisition unit is used to acquire the personal information of the teller to be trained based on the human resources management system.

[0092] Optionally, the error record includes an error code, the number of errors and the percentage of error transactions.

[0093] The device for determining the error training program further described can also execute the method for determining the error training program provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0094] According to an embodiment of the present disclosure, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0095] Figure 4A schematic diagram of the structure of an electronic device 40 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0096] like Figure 4 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0097] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0098] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 41 executes the various methods and processes described above, such as a method for determining an error training scheme.

[0099] In some embodiments, the method for determining an error training scheme may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the method for determining an error training scheme described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to execute the method for determining an error training scheme in any other appropriate manner (e.g., by means of firmware).

[0100] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0102] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0103] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0104] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0105] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0106] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0107] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining an error training program, characterized in that: The method comprises: Obtaining target training reference information of the teller to be trained; the training reference information includes the teller's personal information and the teller's error record in counter operations; Based on the target training reference information, a target training program for this error training is determined through an intelligent judgment model; the target training program includes the training topic of this error training and the tellers to be trained who need to participate in this error training.

2. The method according to claim 1, characterized in that The intelligent judgment model is a trained action strategy network.

3. The method according to claim 2, characterized in that The training process of the action strategy network is as follows: Input the test training reference information of the test teller into the action strategy network to obtain the test training plan for test error training; Executing the test training program, and calculating the function value of the reward function according to the error records of the test teller thereafter; Inputting the function value and the experimental training reference information into the action evaluation network to obtain the action evaluation index; The action strategy network is subjected to intensive training according to the action evaluation index.

4. The method according to claim 3, characterized in that The function value of the reward function is calculated according to the error record of the test teller thereafter, including: Obtain the total number of transactions and total number of errors of the test teller thereafter; The function value of the reward function is calculated according to the ratio of the total number of errors to the total number of transactions.

5. The method according to claim 1, characterized in that The step of obtaining target training reference information of the teller to be trained includes: Obtain error records of the tellers to be trained in counter operations based on the counter system and operation management process; Obtain personal information of tellers to be trained based on the human resources management system.

6. The method according to claim 5, characterized in that The error record includes the error code, the number of errors and the proportion of error transactions.

7. A device for determining an error training program, characterized in that: The device comprises: An acquisition module is used to acquire target training reference information of the teller to be trained; the training reference information includes the teller's personal information and the teller's error record in counter operations; A determination module is used to determine the target training program of this error training based on the target training reference information through an intelligent judgment model; the target training program includes the training topic of this error training and the tellers to be trained who need to participate in this error training.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the method for determining an error training scheme according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining an error training scheme according to any one of claims 1 to 6 when the processor executes the instructions.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the method for determining the error training scheme according to any one of claims 1 to 6.

Images