Business data configuration method and device, storage medium and electronic equipment

Abstract

The invention discloses a service data configuration method and device, a storage medium and electronic equipment. The method relates to the field of artificial intelligence, and comprises the steps that to-be-configured data of a target mechanism is acquired, and the to-be-configured data comprises at least one of mobile service data, static service data, a mobile coverage rate and a yield rate of the static service data; the to-be-configured data are input into a target model, the target adjustment amount of the to-be-configured data is obtained, the target model simulates the environment of the target mechanism, an intelligent agent of the target model is controlled to interact with the environment based on the adjustment amount, and the target adjustment amount with the maximum income is output; and adjusting the to-be-configured data according to the target adjustment amount to obtain target data after the target mechanism is adjusted. Through the service data configuration method and device, the problem that the configuration strategy of the service data is inaccurate due to rapid change of the market environment in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically, to a method, device, storage medium and electronic device for configuring business data. Background Art

[0002] In the daily operations of financial institutions, the allocation ratio of cash to short-term assets is an important part of asset-liability management. Financial institutions need to ensure that they have sufficient liquidity to cope with daily customer withdrawal needs, payment obligations and the maturity of short-term liabilities, while also investing additional funds in the short term to maximize returns. The allocation ratio of cash to short-term investment assets needs to meet corresponding regulatory requirements, such as the liquidity coverage ratio and net stable funding ratio of financial institutions need to meet relevant regulations. These indicators are designed to enable financial institutions to maintain sufficient high-quality liquid assets to cope with potential liquidity risks.

[0003] The difficulty faced by financial institutions in managing short-term liquidity of funds lies in the uncertainty of fund demand. Customers' withdrawal needs, market interest rate fluctuations, loan repayments and other factors will affect the current assets of financial institutions, so financial institutions need to make accurate forecasts of future fund demand and market conditions. In addition, financial institutions need to find a balance between the liquidity security of funds and the return on investment. Holding too many current assets will lead to loss of income, while investing too much in short-term assets may face liquidity risks.

[0004] The asset short-term liquidity management method in the related art is a cash flow forecast and maturity matching strategy based on historical data. However, this method cannot respond to the rapidly changing market and capital demand in real time, especially when there is a sudden liquidity demand or a sharp market fluctuation. The allocation strategy of the current assets and short-term assets of financial institutions is not accurate enough, which makes it difficult to cope with the uncertainty of capital liquidity demand and cannot maximize the efficiency of capital utilization.

[0005] Currently, no effective solution has been proposed to address the problem of inaccurate configuration strategies for business data due to rapid changes in the market environment in related technologies. Summary of the invention

[0006] The main purpose of the present application is to provide a business data configuration method, device, storage medium and electronic device to solve the problem of inaccurate business data configuration strategy due to rapid changes in the market environment in related technologies.

[0007] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for configuring business data is provided. The method comprises: obtaining the data to be configured of the target institution, wherein the data to be configured comprises at least one of the following: mobile business data, static business data, liquidity coverage ratio and rate of return of static business data; inputting the data to be configured into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target institution, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the return; and adjusting the data to be configured according to the target adjustment amount to obtain the adjusted target data of the target institution.

[0008] Optionally, the target model is obtained by: obtaining historical configuration data within a preset period, constructing an environment simulator based on the historical configuration data, and simulating the environment after each interaction between the agent and the environment through the environment simulator, wherein the historical configuration data includes at least one of the following: historical market conditions, historical transaction records, and historical static business data; determining a state space, an action space, a reward function, and an agent, wherein the state space includes state data after each interaction between the agent and the environment, and the state data includes at least one of the following: liquid business data, static business data, liquidity coverage, the rate of return of static business data, and predicted expenditure business data, the action space includes a first adjustment amount of liquid business data and a second adjustment amount of static business data, and the agent includes a policy network and a value network; obtaining interaction data after each interaction through multiple interactions between the agent and the environment, wherein the interaction data includes data in the state space and the action space after each interaction; performing multiple iterative training on the agent based on the interaction data to obtain a target agent, wherein each iterative training adjusts the parameters of the policy network and the value network in the agent, and each iterative training corresponds to an interaction between the agent and the environment; constructing a target model through the environment simulator, the state space, the action space, the reward function, and the target agent.

[0009] Optionally, the environment simulator includes: an input layer, a hidden layer and an output layer, wherein the input layer is used to receive historical configuration data, the hidden layer is used to process the historical configuration data through a preset deep learning algorithm, and the output layer is used to output liquidity coverage ratio, yield of static business data and predicted expenditure business data.

[0010] Optionally, determining the reward function includes: determining a liquidity sufficiency reward item based on the liquidity coverage ratio, determining an investment income reward item based on the yield of static business data, and determining an excessive liquidity penalty item based on the current business data; calculating the sum of the liquidity sufficiency reward item, the investment income reward item and the excessive liquidity penalty item to obtain the reward function.

[0011] Optionally, the agent interacts with the environment multiple times to obtain interaction data after each interaction, including: for each interaction, obtaining state data in the current state space of the environment, and inputting the state data into a policy network in the agent to obtain a first adjustment amount and a second adjustment amount output by the policy network; updating the state data in the state space based on the first adjustment amount and the second adjustment amount to obtain updated state data; calculating a reward value of a reward function based on the updated state data, and determining the first adjustment amount, the second adjustment amount, the state data, the updated state data and the reward value as the interaction data after the interaction.

[0012] Optionally, the intelligent agent is iteratively trained multiple times based on the interaction data to obtain the target intelligent agent, including: for each iterative training, determining the state value function through the value network and interaction data of the intelligent agent, wherein the state value function is used to evaluate the estimated return corresponding to the state data after each interaction; determining the advantage function of the intelligent agent through the state value function, wherein the advantage function is used to evaluate the advantage of the first adjustment amount and the second adjustment amount in each interaction; updating the network parameters of the policy network of the intelligent agent through the advantage function, and updating the network parameters of the value network of the intelligent agent through the estimated return and reward value; when the number of iterative training is greater than or equal to the number threshold, stopping the iterative training, and determining the policy network and value network after the last iterative training as the target intelligent agent.

[0013] Optionally, updating the network parameters of the agent's policy network through the advantage function, and updating the network parameters of the agent's value network through the estimated return and reward value includes: determining the clipping objective function of the policy network, inputting the advantage function into the clipping objective function, and updating the network parameters of the policy network until the function value of the clipping objective function is maximum; inputting the estimated return and reward value into the mean square error formula, and updating the network parameters of the value network until the output value of the mean square error formula is minimum.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a configuration device for business data is provided. The device comprises: an acquisition unit, used to acquire the data to be configured of the target institution, wherein the data to be configured comprises at least one of the following: mobile business data, static business data, liquidity coverage ratio and the rate of return of static business data; an input unit, used to input the data to be configured into the target model, and obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target institution, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the benefit; an adjustment unit, used to adjust the data to be configured according to the target adjustment amount, and obtain the adjusted target data of the target institution.

[0015] In an embodiment of the present application, the data to be configured of the target institution is obtained, wherein the data to be configured includes at least one of the following: mobile business data, static business data, liquidity coverage ratio and rate of return of static business data; the data to be configured is input into the target model to obtain a target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target institution, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs a target adjustment amount that maximizes the profit; the data to be configured is adjusted according to the target adjustment amount to obtain the adjusted target data of the target institution, and the target model is trained by adopting a deep reinforcement learning algorithm, and the market environment faced by the target institution is simulated. Through the interaction between the intelligent agent and the environment, the target adjustment amount that maximizes the profit is learned, and the data to be configured is adjusted based on the target adjustment amount, thereby achieving a balance between the security and profitability of the business data, optimizing the configuration strategy, and thus achieving the technical effect of improving the accuracy of the configuration strategy of the business data, thereby solving the technical problem of inaccurate configuration strategy of the business data due to the rapid changes in the market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for configuring business data is shown;

[0018] Figure 2 is a flowchart of a method for configuring business data provided in an embodiment of the present application;

[0019] Figure 3 is a schematic diagram of an updated intelligent agent provided according to an embodiment of the present application;

[0020] Figure 4 is a schematic diagram of an optional method for configuring service data provided according to an embodiment of the present application;

[0021] Figure 5 is a schematic diagram of a device for configuring business data provided in an embodiment of the present application;

[0022] Figure 6 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising 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.

[0025] It should be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0026] Example 1

[0027] According to an embodiment of the present application, a method embodiment for configuring business data is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1The hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for configuring business data is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (shown as 102a, 102b, ..., 102n in the figure) processors 102 (the processor 102 may include but is not limited to a processing device such as an MCU (Microcontroller Unit) or an FPGA (Field-Programmable Gate Array), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a USB (Universal Serial Bus) port (which may be included as one of the ports of a BUS (Business)), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0029] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0030] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the configuration method of business data in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the configuration method of the business data described above is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] The display may be, for example, a touch screen liquid crystal display that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0033] In the above operating environment, Figure 2 is a flowchart of a method for configuring business data according to an embodiment of the present application, such as Figure 2 As shown, the method includes:

[0034] Step S201, obtaining the data to be configured of the target institution, wherein the data to be configured includes at least one of the following: current business data, static business data, liquidity coverage ratio and rate of return of static business data.

[0035] In step S201, the target institution may be a financial institution, the current business data may be data that is allowed to be freely spent or used without restrictions, and the static business data may be data that is restricted and cannot be freely spent or used within a certain period. In some optional examples, the current business data may be the target institution's liquidity, such as cash position, including customer deposits, withdrawals, loan repayments, investment redemptions, etc. The static business data may be short-term assets of investment, such as bonds, short-term held stocks, funds, etc.

[0036] The liquidity coverage ratio is the ratio of high-quality liquid assets held by a financial institution to its expected net cash outflow in the next 30 days, which is used to indicate the ability of a financial institution to meet short-term liquidity needs. The yield of static business data can be the average or specific yield of a financial institution's short-term assets.

[0037] Step S202, inputting the data to be configured into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target organization, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the benefit.

[0038] In step S202, the target model can be a machine learning model using a reinforcement learning algorithm, and the real business environment and market conditions of the target institution (such as a financial institution) are simulated through the environment simulator in the target model. The environment simulator may include: state space, action space, reward function and transition model, wherein the state space is defined by the data to be configured, and may include flow business data, static business data, liquidity coverage and the rate of return of static business data, etc. The action space represents the adjustment actions that the intelligent agent of the target model can take, such as increasing or decreasing flow business data and static business data. The reward function is defined based on the target institution's profit maximization and liquidity management goals, such as using liquidity sufficiency rewards, investment income rewards and excessive liquidity penalties to determine the reward function. The transition model is used to describe the rules of state change, that is, how the state of the environment changes after the intelligent agent takes action.

[0039] The agent is a learning entity in the reinforcement learning model, which is used to observe the state of the environment in the environment simulated by the reinforcement learning model, take actions, and optimize the action strategy based on the rewards fed back by the environment. The acquired data to be configured is input into the target model as the starting state. The agent in the target model outputs a target adjustment amount based on the current state and its policy network as its action at the current time step. The environment simulates state changes based on the actions of the agent and the built-in transition model, and generates new states and immediate rewards. The agent receives feedback from the environment (i.e. rewards) and updates its strategy (i.e. adjustments) according to the advantage function so that future decisions in similar states can obtain higher benefits.

[0040] The interaction between the agent and the environment is a multi-round iterative process, each of which includes the agent taking actions, the environment responding to state changes, and reward feedback. The goal of the agent is to maximize the cumulative reward (that is, maximize the return) in a long-term time frame, that is, to maximize investment returns and minimize liquidity holding costs while meeting liquidity coverage requirements. After a series of iterative training, the agent in the target model will learn how to maximize the target adjustment amount of long-term cumulative rewards based on its policy network output within a given state space.

[0041] Step S203, adjusting the data to be configured according to the target adjustment amount to obtain the target data after the target mechanism is adjusted.

[0042] In step S203, after obtaining the target adjustment amount, the data to be configured is adjusted according to the target adjustment amount to obtain the target data. For example, the data to be configured includes mobile business data and static business data, and the target adjustment amount includes a first adjustment amount for the mobile business data and a second adjustment amount for the static business data. The mobile business data is adjusted by the first adjustment amount, and the static business data is adjusted by the second adjustment amount, so that the target institution can maximize investment returns and reduce the retention cost of mobile business data while meeting the liquidity coverage requirements.

[0043] In some examples, the target adjustment amount can be a vector or structured data of financial parameter adjustment recommendations such as the amount of cash reserves and the amount of short-term asset investment of a financial institution. According to the recommendations on cash reserves in the target adjustment amount, the financial institution needs to adjust its cash holdings. For example, if the agent recommends increasing cash reserves, the financial institution needs to redeem funds from short-term assets or raise cash through other channels to ensure sufficient liquidity to meet customer needs and deal with potential liquidity risks. For short-term asset investments, the agent may recommend adjusting the investment portfolio to pursue higher returns or adjusting the risk level according to changes in market conditions. Financial institutions need to reallocate their investment portfolios based on the recommendations in the target adjustment amount, such as buying or selling specific short-term assets, to achieve better capital allocation.

[0044] After adjusting the cash and short-term asset allocation, the liquidity coverage ratio and the rate of return on short-term assets need to be recalculated to reflect changes in the actual business environment. The liquidity coverage ratio needs to ensure that financial institutions meet the regulatory requirements of relevant departments, while the update of the rate of return reflects the expected return level of the target data.

[0045] The business data configuration method provided in the embodiment of the present application obtains the data to be configured of the target organization, wherein the data to be configured includes at least one of the following: mobile business data, static business data, liquidity coverage ratio and rate of return of static business data; inputs the data to be configured into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target organization, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the profit; adjusts the data to be configured according to the target adjustment amount to obtain the adjusted target data of the target organization, trains the target model by adopting a deep reinforcement learning algorithm, simulates the market environment faced by the target organization, learns the target adjustment amount that maximizes the profit through the interaction between the intelligent agent and the environment, and adjusts the data to be configured based on the target adjustment amount, thereby achieving a balance between the security and profitability of the business data and optimizing the configuration strategy, thereby achieving the technical effect of improving the accuracy of the configuration strategy of the business data, and further solving the technical problem of inaccurate configuration strategy of the business data due to the rapid changes in the market environment.

[0046] In order to accurately predict the target adjustment amount of the data to be configured, it is necessary to train a target model. Optionally, in the configuration method of business data provided in the embodiment of the present application, the target model is obtained by: obtaining historical configuration data within a preset period, building an environment simulator based on the historical configuration data, and simulating the environment after each interaction between the agent and the environment through the environment simulator, wherein the historical configuration data includes at least one of the following: historical market conditions, historical transaction records, and historical static business data; determining a state space, an action space, a reward function, and an agent, wherein the state space includes state data after each interaction between the agent and the environment, and the state data includes at least one of the following: flowing business data, static business data, flow business data, dynamic coverage, yield of static business data and predicted expenditure business data, the action space includes a first adjustment amount of the flow business data and a second adjustment amount of the static business data, and the intelligent agent includes a policy network and a value network; the intelligent agent interacts with the environment for multiple times to obtain interaction data after each interaction, wherein the interaction data includes data in the state space and action space after each interaction; the intelligent agent is iteratively trained for multiple times based on the interaction data to obtain a target intelligent agent, wherein each iterative training adjusts the parameters of the policy network and the value network in the intelligent agent, and each iterative training corresponds to an interaction between the intelligent agent and the environment; a target model is constructed through an environment simulator, a state space, an action space, a reward function and a target intelligent agent.

[0047] In some examples, historical configuration data within a preset period is collected from the historical records of the target institution. This historical configuration data may include historical market conditions (such as interest rates, stock market performance, etc.), historical transaction records (such as cash inflow and outflow records, short-term asset purchase and sale records, etc.), and historical static business data (such as long-term investment portfolios, fixed costs, etc.). Based on the historical configuration data, an environmental simulator is constructed, which can reflect market changes, business operations, and capital flows. The environmental simulator provides a virtual scene for each interaction between the agent and the environment, enabling the agent to learn and optimize strategies without interfering with actual business.

[0048] The state space is used to define the observable state of the agent at each decision point (adjustment), including but not limited to current business data, static business data, liquidity coverage ratio, rate of return of static business data, and predicted expenditure business data. These states describe the current financial status and market environment of the financial institution. The action space is used to define the range of decisions that the agent can take, including the first adjustment of current business data (such as the increase or decrease of cash reserves) and the second adjustment of static business data (such as the adjustment of short-term asset investment). For example, the action recorded in the action space is A t To represent, where A t ={ΔC t ∈[-ΔCmax , ΔC max ], ΔI t ∈[-ΔI max , ΔI max ]}, where ΔC max Expressed as the maximum adjustment of cash reserves, ΔI max is the maximum adjustment of short-term assets, and 0 means no adjustment. The agent will take actions in the environment based on these adjustments. The reward function is used to quantify the quality of the agent's decision. The reward function can include liquidity sufficiency rewards, that is, the requirements of liquidity coverage ratio, investment returns, that is, the returns of short-term assets, and penalties for excessive liquidity, that is, the opportunity cost of too much cash reserves, to guide the agent to balance liquidity risk and maximize returns.

[0049] The agent can include a policy network and a value network. The policy network is responsible for selecting the best action based on the current state, while the value network evaluates the long-term value of the state. In the environment simulator, the agent interacts with the environment multiple times and collects the interaction data of the state, action, reward, and next state after each interaction. These data constitute the learning samples for training and optimizing the agent strategy. Using the collected interaction data, the policy network and value network in the agent are trained multiple times iteratively, with the goal of maximizing the cumulative reward while limiting the amplitude of the policy update to ensure the stability of the training and the coherence of the strategy.

[0050] After sufficient training, the policy network and value network learned by the agent can predict the optimal action and state value based on the current state data. The environment simulator, state space, action space, reward function and trained target agent are combined to form the final target model, which can output the optimal cash and short-term asset allocation recommendations based on market conditions and financial institutions' business data.

[0051] This embodiment trains the target model so that the target model can simulate the capital allocation decision of financial institutions under market fluctuations. The strategy network learned by the agent in the model can guide financial institutions to adjust their cash reserves and short-term asset investments in real time or regularly to achieve the goals of sufficient liquidity, maximized returns and reduced opportunity costs. The target model can adjust the allocation strategy of cash and short-term assets in real time and dynamically, allowing financial institutions to make more accurate capital allocation decisions based on future market fluctuations and capital needs. The flexibility and adaptability of capital allocation are significantly improved.

[0052] In the target model, an environment simulator is needed to simulate the environment of the target organization. Optionally, in the business data configuration method provided in the embodiment of the present application, the environment simulator includes: an input layer, a hidden layer and an output layer, wherein the input layer is used to receive historical configuration data, the hidden layer is used to process the historical configuration data through a preset deep learning algorithm, and the output layer is used to output liquidity coverage, the yield of static business data and predicted expenditure business data.

[0053] In some examples, the input layer of the environment simulator receives historical configuration data and preprocesses the historical configuration data, such as normalization, filling missing values, outlier detection, etc. The hidden layer can process the input data using a preset deep learning algorithm, which includes but is not limited to a convolutional neural network, a long short-term memory network, a gated recurrent unit, or a self-attention mechanism. The hidden layer extracts features from the data preprocessed by the input layer and establishes a complex relationship between these features and future prediction indicators. The output layer is used to generate three key prediction indicators: liquidity coverage ratio, the yield of static business data, and predicted expenditure business data. The prediction of liquidity coverage ratio helps financial institutions estimate future liquidity conditions and the ability to meet regulatory requirements. The prediction of yield provides expectations about the return on asset investment and guides capital allocation decisions to pursue maximum returns. Predicted expenditure business data helps financial institutions estimate future funding needs, such as customer withdrawals, loan repayments, and other expenditures.

[0054] This embodiment uses an environment simulator to simulate possible future state changes of the environment, providing a virtual environment for the agent to learn and optimize. This allows the agent to learn how to make optimal fund allocation decisions under different market conditions without actual market fluctuations, thereby achieving more efficient and robust liquidity management in practical applications. This significantly improves the efficiency and response speed of short-term liquidity management of funds.

[0055] A reward function is used to evaluate the quality of each action of the intelligent agent. Optionally, in the method for configuring business data provided in the embodiment of the present application, determining the reward function includes: determining a liquidity sufficiency reward item based on the liquidity coverage ratio, determining an investment income reward item based on the rate of return of static business data, and determining an excessive liquidity penalty item based on the flowing business data; calculating the sum of the liquidity sufficiency reward item, the investment income reward item and the excessive liquidity penalty item to obtain the reward function.

[0056] In some examples, the calculation formula of the reward function may be as follows:

[0057]

[0058] in, As a reward for sufficient liquidity, For investment income incentives, is the excessive liquidity penalty term. The sufficient liquidity reward term can be expressed by the following formula:

[0059]

[0060] If the liquidity coverage ratio LCR of the day t Meet the regulatory requirements of relevant institutions (set as LCR target ), a positive reward is given; otherwise, a penalty is given. Among them, α1 is the positive reward when the liquidity requirement is met, and α2 is the penalty coefficient, which increases with the increase of liquidity deficiency. The investment income reward item can be expressed by the following formula:

[0061]

[0062] Among them, β is the reward coefficient of investment income. The short-term assets invested by financial institutions bring daily income, r t is the rate of return of short-term assets on day t, I t is the amount of short-term assets held on that day. Investment income rewards are proportional to daily income. The excessive liquidity penalty term can be expressed by the following formula:

[0063]

[0064] Among them, γ is the penalty coefficient for excess liquidity; if C t ≤C max ,but Although excessive cash reserves can improve liquidity, they will incur opportunity costs due to the inability to obtain investment returns. t is the cash reserve on day t. If the cash reserve exceeds the upper limit C set by the financial institution, max , then punishment needs to be imposed.

[0065] This embodiment determines a reward function to evaluate the quality of each action selected by the agent, thereby helping financial institutions to select more accurate allocation strategies for capital allocation strategies.

[0066] In order to optimize the intelligent agent, it is necessary to obtain the interaction data between the intelligent agent and the environment. Optionally, in the business data configuration method provided in the embodiment of the present application, the intelligent agent interacts with the environment multiple times to obtain the interaction data after each interaction, including: for each interaction, obtaining the state data in the current state space of the environment, and inputting the state data into the policy network in the intelligent agent to obtain the first adjustment amount and the second adjustment amount output by the policy network; updating the state data in the state space based on the first adjustment amount and the second adjustment amount to obtain the updated state data; calculating the reward value of the reward function based on the updated state data, and determining the first adjustment amount, the second adjustment amount, the state data, the updated state data and the reward value as the interaction data after the interaction.

[0067] In some examples, at the beginning of each interaction, the environment provides the agent with current state data. The state data reflects the financial status and market conditions of the target institution, and may include current business data, static business data, liquidity coverage ratio, rate of return of static business data, and predicted expenditure business data. The agent takes the received state data as input and passes it to its internal policy network. The policy network outputs the action that the agent should take based on the current state and the learned policy, i.e., the first adjustment amount of the current business data and the second adjustment amount of the static business data. For example, specific operations such as increase or decrease of cash reserves and adjustment of short-term asset investment.

[0068] The action taken by the agent affects the state of the environment, that is, the environment simulator updates its internal state space according to the first adjustment amount and the second adjustment amount, and generates updated state data. The environment calculates the reward value of the reward function based on the action of the agent and the updated state data. The reward value reflects the impact of the action on the fund management efficiency and risk control of the target institution, including liquidity sufficiency rewards, investment income rewards and penalties for excessive cash retention.

[0069] The first adjustment amount taken by the agent, the second adjustment amount, the original state data of the environment, the updated state data, and the calculated reward value together constitute the interaction data after an interaction. These data are collected and stored by the agent for subsequent model training and strategy optimization. Through the interaction between the agent and the environment, it is repeated in multiple time steps. The data of each interaction is accumulated to build an experience replay pool, which is the basis for the training of deep reinforcement learning algorithms.

[0070] This embodiment continuously learns and optimizes the agent's capital allocation strategy by acquiring the interaction data between the agent and the environment as training data, so that the agent can make better decisions in a complex and changing market environment to achieve the optimization of the target organization's business data management.

[0071] After acquiring the interaction data, the agent is iteratively trained using the interaction data. Optionally, in the method for configuring business data provided in an embodiment of the present application, the agent is iteratively trained multiple times based on the interaction data to obtain a target agent, including: for each iterative training, determining a state value function through the agent's value network and interaction data, wherein the state value function is used to evaluate the estimated return corresponding to the state data after each interaction; determining the agent's advantage function through the state value function, wherein the advantage function is used to evaluate the advantage of the first adjustment amount and the second adjustment amount in each interaction; updating the network parameters of the agent's policy network through the advantage function, and updating the network parameters of the agent's value network through the estimated return and reward value; when the number of iterative training is greater than or equal to a threshold number, stopping the iterative training, and determining the policy network and value network after the last iterative training as the target agent.

[0072] In some examples, the state data in the collected interaction data are evaluated using the agent's value network to calculate a state value function, denoted as V(s). The state value function V(s) reflects the average future return that the agent can expect to obtain based on the current strategy in state s. The advantage function measures the difference between the reward of taking action a in state s and the average reward of taking any action in state s. The advantage function is calculated based on the state value function V(s) and the action value function Q(s,a). Q(s,a) is the future return that the agent can expect to obtain based on the current strategy after taking action a in state s.

[0073] The calculation formula of the advantage function can be:

[0074] A(s,a)=Q(s,a)-V(s);

[0075] Where A(s,a) represents the advantage of taking action a in state s.

[0076] The update of the policy network is based on the advantage function A(s,a), and the probability distribution of the action taken by the agent is optimized by the gradient ascent method. The parameters of the policy network are adjusted to increase the probability of taking high-advantage actions in beneficial states. In this embodiment, the clipping objective function in the PPO algorithm (Proximal Policy Optimization, an algorithm for optimizing reinforcement learning strategies) can be used to update the policy network. The function takes into account the ratio of the new and old strategies, limits the update range of the strategy, and ensures the stability of the training and the continuity of the strategy.

[0077] The value network is updated based on the estimated rewards and the immediate reward values ​​in the interaction data. The estimated rewards are based on the dynamic evaluation of the agent's strategy and the environment, reflecting the long-term benefits that the agent can expect to obtain after taking a series of actions. The parameters of the value network are updated using the mean squared error as the loss function to minimize the difference between the predicted value and the actual reward. This helps the value network more accurately assess the long-term value of each state.

[0078] During the iterative training process, a number threshold N can be set. When the number of iterative training reaches or exceeds N, the training process is stopped. This threshold can be set based on experiments to balance training efficiency and model performance. After the last iterative training, the policy network and value network parameters of the agent are updated to obtain the target agent.

[0079] For example, Figure 3 is a schematic diagram of updating an intelligent agent according to an embodiment of the present application, such as Figure 3 As shown in the figure, the original state data of each interaction between the bank and the market environment, as well as the state data and rewards after the interaction, are collected. The action of adjusting cash and short-term asset reserves is performed through the actor network (also known as the policy network), and the long-term value of the state of each interaction is evaluated through the critic network (also known as the value network). The PPO algorithm is used to update the actor network and the critic network. For the actor network, a policy cutting mechanism is introduced to limit the range of changes between the new and old strategies, and the adjustment strategy of each interaction of the actor network is updated by maximizing the advantage function. For the critic network, the parameters of the critic network are optimized by minimizing the mean square error between the estimated return and the reward value.

[0080] The PPO algorithm is used to automatically learn the optimal strategy. The algorithm adjusts the strategy update amplitude through "clipping" to control the size of each strategy change. Specifically: PPO limits the amplitude of the strategy update to ensure that each update does not deviate too far from the original strategy. This constraint avoids the problem of unstable training caused by excessive strategy changes. At each update, PPO will adjust the step size by comparing the strategies before and after the update to ensure that the model can robustly converge to the optimal solution.

[0081] This embodiment iteratively trains the agent through interactive data. The value network of the agent learns the ability to evaluate the state value, while the policy network learns the strategy of taking the best action under different states. This process enables the agent to dynamically adjust the cash reserve and short-term asset investment amount, etc. in the environment of a simulated financial institution based on state information such as current business data, static business data, liquidity coverage, the rate of return of static business data, and predicted expenditure business data, so as to achieve the best balance between sufficient liquidity and maximum investment income. The target agent finally obtained can output the optimal fund allocation decision according to the actual situation, thereby improving the bank's fund management efficiency and risk control capabilities.

[0082] Optionally, in the method for configuring business data provided in an embodiment of the present application, updating the network parameters of the agent's policy network through the advantage function, and updating the network parameters of the agent's value network through the estimated return and reward value include: determining the trimming objective function of the policy network, inputting the advantage function into the trimming objective function, and updating the network parameters of the policy network until the function value of the trimming objective function is maximum; inputting the estimated return and reward value into the mean square error formula, and updating the network parameters of the value network until the output value of the mean square error formula is minimum.

[0083] In some examples, the clipping objective function may be as follows:

[0084]

[0085] in, is the ratio of the new and old strategies, and ε is the clipping amplitude. Clip represents the clipping function, and the clipping objective function ensures that the update of the policy network will not "jump" too much. t represents the advantage function, Express expectations.

[0086] Through gradient descent or other optimization algorithms, the function value of the clipping objective function is maximized and the parameters of the policy network are updated. This process can be completed through the back propagation algorithm, adjusting the weights and biases in the policy network to maximize the value of the clipping objective function, thereby optimizing the decision-making ability of the policy network.

[0087] The update of the value network is based on the estimated return (i.e., the estimate of the state value function) and the actual immediate reward received. The estimated return can be obtained by weighted average of the immediate reward and the future discounted return. The value network can use the mean square error as the loss function, where the loss function of the value network can be expressed by the following formula:

[0088] L VF (θ) = E t [(V θ (s t)-R t )] 2 ;

[0089] Among them, V θ (s t ) is the value of the predicted state value function, i.e., the estimated return, R t is the actual return of the current state, that is, the reward value. Through the gradient descent algorithm, the mean square error loss function is minimized and the parameters of the value network are updated.

[0090] This embodiment updates the policy network and the value network, so that the agent can gradually optimize its strategy in the environment in multiple iterations, achieving more efficient fund management and risk control. The minimization of the clipping objective function and the mean square error loss function ensures that the learning process of the policy network and the value network can converge quickly while remaining robust and continuous, so that the agent can make more reasonable and optimized decisions in the complex and changing financial market environment.

[0091] According to another embodiment of the present application, an optional method for configuring service data is also provided. Figure 4 is a schematic diagram of an optional method for configuring service data according to an embodiment of the present application, such as Figure 4 As shown in the figure, historical market conditions and bank liquidity management related data are collected and preprocessed. A dynamic optimization model for bank cash and short-term asset allocation is established and a reward function is constructed. The model is trained by combining the data set, environment simulator and reward function. The trained model is used to obtain the optimal allocation plan for bank short-term liquidity.

[0092] This embodiment uses an optional business data configuration method, combined with deep learning predictions of market conditions, cash outflows, and short-term asset returns, to dynamically adjust the cash and short-term asset allocation strategy in real time, allowing banks to make more accurate fund allocation decisions based on future market fluctuations and funding needs, significantly improving the flexibility and adaptability of fund allocation.

[0093] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0094] Example 2

[0095] The embodiment of the present application also provides a device for configuring business data. It should be noted that the device for configuring business data in the embodiment of the present application can be used to execute the method for configuring business data provided in the embodiment of the present application. The device for configuring business data provided in the embodiment of the present application is introduced below.

[0096] According to an embodiment of the present application, a device for implementing the above-mentioned service data configuration method is also provided. Figure 5 is a schematic diagram of a configuration device for service data provided in an embodiment of the present application, such as Figure 5 As shown, the device comprises:

[0097] The acquisition unit 501 is used to acquire the data to be configured of the target institution, wherein the data to be configured includes at least one of the following: current business data, static business data, liquidity coverage ratio and rate of return of static business data;

[0098] An input unit 502 is used to input the data to be configured into the target model to obtain a target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target mechanism, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the benefit;

[0099] The adjustment unit 503 is used to adjust the data to be configured according to the target adjustment amount to obtain the target data after the target mechanism is adjusted.

[0100] The business data configuration device provided in the embodiment of the present application obtains the data to be configured of the target organization through the acquisition unit 501, wherein the data to be configured includes at least one of the following: mobile business data, static business data, liquidity coverage ratio and the rate of return of static business data; the input unit 502 inputs the data to be configured into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target organization, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the profit; the adjustment unit 503 adjusts the data to be configured according to the target adjustment amount to obtain the adjusted target data of the target organization, trains the target model by adopting a deep reinforcement learning algorithm, simulates the market environment faced by the target organization, learns the target adjustment amount that maximizes the profit through the interaction between the intelligent agent and the environment, and adjusts the data to be configured based on the target adjustment amount, thereby achieving a balance between the security and profitability of the business data, optimizing the configuration strategy, and thus achieving the technical effect of improving the accuracy of the configuration strategy of the business data, thereby solving the technical problem of inaccurate configuration strategy of the business data due to the rapid changes in the market environment.

[0101] Optionally, in the business data configuration device provided in the embodiment of the present application, the device also includes: a first construction unit, used to obtain historical configuration data within a preset period, build an environment simulator based on the historical configuration data, and simulate the environment after each interaction between the agent and the environment through the environment simulator, wherein the historical configuration data includes at least one of the following: historical market conditions, historical transaction records, and historical static business data; a first determination unit, used to determine the state space, action space, reward function, and agent, wherein the state space includes state data after each interaction between the agent and the environment, and the state data includes at least one of the following: liquid business data, static business data, liquidity coverage, rate of return of static business data, and predicted support The invention discloses a method for preparing a target intelligent agent and a target agent for processing business data, wherein the action space includes a first adjustment amount for the flow business data and a second adjustment amount for the static business data, and the intelligent agent includes a policy network and a value network; an interaction unit is used to perform multiple interactions between the intelligent agent and the environment to obtain interaction data after each interaction, wherein the interaction data includes data in the state space and the action space after each interaction; a training unit is used to perform multiple iterative training on the intelligent agent based on the interaction data to obtain a target intelligent agent, wherein each iterative training adjusts the parameters of the policy network and the value network in the intelligent agent, and each iterative training corresponds to an interaction between the intelligent agent and the environment; a second construction unit is used to construct a target model through an environment simulator, a state space, an action space, a reward function and a target intelligent agent.

[0102] Optionally, in the business data configuration device provided in the embodiment of the present application, the environment simulator includes: an input layer, a hidden layer and an output layer, wherein the input layer is used to receive historical configuration data, the hidden layer is used to process the historical configuration data through a preset deep learning algorithm, and the output layer is used to output liquidity coverage ratio, the yield of static business data and predicted expenditure business data.

[0103] Optionally, in the business data configuration device provided in the embodiment of the present application, the first determination unit includes: a first determination module, used to determine the liquidity sufficiency reward item based on the liquidity coverage ratio, determine the investment income reward item based on the yield of static business data, and determine the excessive liquidity penalty item based on the current business data; a first calculation module, used to calculate the sum of the liquidity sufficiency reward item, the investment income reward item and the excessive liquidity penalty item to obtain a reward function.

[0104] Optionally, in the configuration device for business data provided in the embodiment of the present application, the interaction unit includes: an acquisition module, which is used to acquire the state data in the current state space of the environment for each interaction, and input the state data into the policy network in the intelligent agent to obtain a first adjustment amount and a second adjustment amount output by the policy network; a first update module, which is used to update the state data in the state space based on the first adjustment amount and the second adjustment amount to obtain updated state data; a second calculation module, which is used to calculate the reward value of the reward function based on the updated state data, and determine the first adjustment amount, the second adjustment amount, the state data, the updated state data and the reward value as the interaction data after the interaction.

[0105] Optionally, in the configuration device of business data provided in the embodiment of the present application, the training unit includes: a second determination module, used to determine the state value function through the value network and interaction data of the agent for each iterative training, wherein the state value function is used to evaluate the estimated return corresponding to the state data after each interaction; a third determination module, used to determine the advantage function of the agent through the state value function, wherein the advantage function is used to evaluate the advantage of the first adjustment amount and the second adjustment amount in each interaction; a second update module, used to update the network parameters of the policy network of the agent through the advantage function, and update the network parameters of the value network of the agent through the estimated return and reward value; a fourth determination module, used to stop the iterative training when the number of iterative training is greater than or equal to the number threshold, and determine the policy network and value network after the last iterative training as the target agent.

[0106] Optionally, in the configuration device for business data provided in the embodiment of the present application, the second update module includes: a determination submodule, used to determine the trimming objective function of the policy network, input the advantage function into the trimming objective function, and update the network parameters of the policy network until the function value of the trimming objective function is maximum; an input submodule, used to input the estimated return and reward value into the mean square error formula, and update the network parameters of the value network until the output value of the mean square error formula is minimum.

[0107] It should be noted that the acquisition unit 501, input unit 502 and adjustment unit 503 correspond to steps S201 to S203 in Example 1, and the two modules and the corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the above-mentioned modules may also be part of the device and may be run in the computer terminal 10 provided in Example 1.

[0108] Example 3

[0109] An embodiment of the present application may provide an electronic device, Figure 6 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 6 As shown, the electronic device may include: one or more ( Figure 6 (only one is shown) processor 602, memory 604, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0110] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the data to be configured of the target institution, wherein the data to be configured includes at least one of the following: mobile business data, static business data, liquidity coverage ratio and the rate of return of static business data; input the data to be configured into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target institution, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the profit; adjusts the data to be configured according to the target adjustment amount to obtain the adjusted target data of the target institution.

[0112] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain historical configuration data within a preset period, build an environment simulator based on the historical configuration data, and simulate the environment after each interaction between the agent and the environment through the environment simulator, wherein the historical configuration data includes at least one of the following: historical market conditions, historical transaction records, and historical static business data; determine the state space, action space, reward function, and agent, wherein the state space includes state data after each interaction between the agent and the environment, and the state data includes at least one of the following: liquid business data, static business data, liquidity coverage, and the income of static business data. rate and predicted expenditure business data, the action space includes a first adjustment amount for the flow business data and a second adjustment amount for the static business data, the intelligent agent includes a policy network and a value network; the intelligent agent interacts with the environment for multiple times to obtain interaction data after each interaction, wherein the interaction data includes data in the state space and the action space after each interaction; the intelligent agent is iteratively trained for multiple times based on the interaction data to obtain a target intelligent agent, wherein each iterative training adjusts the parameters of the policy network and the value network in the intelligent agent, and each iterative training corresponds to an interaction between the intelligent agent and the environment; a target model is constructed through an environment simulator, a state space, an action space, a reward function and a target intelligent agent.

[0113] The processor can also call the information and applications stored in the memory through the transmission device to perform the following steps: The environment simulator includes: an input layer, a hidden layer and an output layer, wherein the input layer is used to receive historical configuration data, the hidden layer is used to process the historical configuration data through a preset deep learning algorithm, and the output layer is used to output liquidity coverage, the yield of static business data and predicted expenditure business data.

[0114] The processor can also call the information and applications stored in the memory through the transmission device to perform the following steps: determine the liquidity sufficiency reward item based on the liquidity coverage ratio, determine the investment income reward item based on the yield of static business data, and determine the excessive liquidity penalty item based on the current business data; calculate the sum of the liquidity sufficiency reward item, the investment income reward item and the excessive liquidity penalty item to obtain the reward function.

[0115] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: for each interaction, obtain the state data in the current state space of the environment, and input the state data into the policy network in the intelligent agent to obtain the first adjustment amount and the second adjustment amount output by the policy network; update the state data in the state space based on the first adjustment amount and the second adjustment amount to obtain the updated state data; calculate the reward value of the reward function based on the updated state data, and determine the first adjustment amount, the second adjustment amount, the state data, the updated state data and the reward value as the interaction data after the interaction.

[0116] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: for each iterative training, determine the state value function through the value network and interaction data of the intelligent agent, wherein the state value function is used to evaluate the estimated return corresponding to the state data after each interaction; determine the advantage function of the intelligent agent through the state value function, wherein the advantage function is used to evaluate the advantage of the first adjustment amount and the second adjustment amount in each interaction; update the network parameters of the policy network of the intelligent agent through the advantage function, and update the network parameters of the value network of the intelligent agent through the estimated return and reward value; when the number of iterative training is greater than or equal to the number threshold, stop the iterative training, and determine the policy network and value network after the last iterative training as the target intelligent agent.

[0117] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: determine the trimming objective function of the policy network, input the advantage function into the trimming objective function, and update the network parameters of the policy network until the function value of the trimming objective function is the maximum value; input the estimated return and reward value into the mean square error formula, and update the network parameters of the value network until the output value of the mean square error formula is the minimum value.

[0118] By adopting the embodiment of the present application, a method is provided to obtain the data to be configured of the target institution, wherein the data to be configured includes at least one of the following: mobile business data, static business data, liquidity coverage ratio and the rate of return of static business data; the data to be configured is input into the target model to obtain the target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target institution, controls the intelligent agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the profit; adjusts the data to be configured according to the target adjustment amount, and obtains the solution of the adjusted target data of the target institution. By adopting a deep reinforcement learning algorithm to train the target model, simulating the market environment faced by the target institution, learning the target adjustment amount to maximize the profit through the interaction between the intelligent agent and the environment, and adjusting the data to be configured based on the target adjustment amount, the security and profitability of the business data are balanced, the configuration strategy is optimized, and the technical effect of improving the accuracy of the configuration strategy of the business data is achieved, thereby solving the technical problem of inaccurate configuration strategy of the business data due to the rapid changes in the market environment.

[0119] It can be understood by those skilled in the art that Figure 6 The structure shown is for illustration only, and the electronic device may also be a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), a PAD or other terminal device. Figure 6The structure of the electronic device is not limited. Figure 6 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 6 Different configurations shown.

[0120] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0121] Example 4

[0122] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the method for configuring the service data provided in the first embodiment.

[0123] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0124] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the method for configuring business data.

[0125] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0126] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, 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 interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0128] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] In addition, each functional unit in each embodiment of the present application 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0131] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for configuring business data, characterized in that: include: Acquire the data to be configured of the target institution, wherein the data to be configured includes at least one of the following: current business data, static business data, liquidity coverage ratio and the rate of return of the static business data; Inputting the data to be configured into a target model to obtain a target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target mechanism, controls the agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the benefit; The data to be configured is adjusted according to the target adjustment amount to obtain the target data after the target mechanism is adjusted.

2. The method according to claim 1, characterized in that The target model is obtained by: Acquire historical configuration data within a preset period, build an environment simulator based on the historical configuration data, and simulate the environment after each interaction between the agent and the environment through the environment simulator, wherein the historical configuration data includes at least one of the following: historical market conditions, historical transaction records, and historical static business data; Determine a state space, an action space, a reward function and an agent, wherein the state space includes state data after each interaction between the agent and the environment, the state data includes at least one of the following: flowing business data, static business data, liquidity coverage, the rate of return of the static business data and predicted expenditure business data, the action space includes a first adjustment amount of the flowing business data and a second adjustment amount of the static business data, and the agent includes a policy network and a value network; The agent interacts with the environment multiple times to obtain interaction data after each interaction, wherein the interaction data includes data in the state space and the action space after each interaction; Performing multiple iterations of training on the agent based on the interaction data to obtain a target agent, wherein each iteration of training adjusts the parameters of the policy network and the value network in the agent, and each iteration of training corresponds to an interaction between the agent and the environment; The target model is constructed by the environment simulator, the state space, the action space, the reward function and the target agent.

3. The method according to claim 2, characterized in that The environmental simulator includes: an input layer, a hidden layer and an output layer, wherein the input layer is used to receive the historical configuration data, the hidden layer is used to process the historical configuration data through a preset deep learning algorithm, and the output layer is used to output the liquidity coverage ratio, the yield of the static business data and the predicted expenditure business data.

4. The method according to claim 2, characterized in that: Determining the reward function includes: Determine the liquidity sufficiency reward item based on the liquidity coverage ratio, determine the investment income reward item based on the yield of static business data, and determine the excessive liquidity penalty item based on the flow business data; The reward function is obtained by calculating the sum of the sufficient liquidity reward item, the investment income reward item and the excessive liquidity penalty item.

5. The method according to claim 2, characterized in that: The agent interacts with the environment multiple times, and the interaction data obtained after each interaction includes: For each interaction, obtaining state data in the current state space of the environment, and inputting the state data into a policy network in the agent to obtain a first adjustment amount and a second adjustment amount output by the policy network; updating the state data in the state space based on the first adjustment amount and the second adjustment amount to obtain updated state data; A reward value of the reward function is calculated based on the updated state data, and the first adjustment amount, the second adjustment amount, the state data, the updated state data, and the reward value are determined as interaction data after interaction.

6. The method according to claim 2, characterized in that The agent is trained iteratively multiple times based on the interaction data to obtain a target agent including: For each iteration of training, a state value function is determined by the value network of the agent and the interaction data, wherein the state value function is used to evaluate the estimated return corresponding to the state data after each interaction; Determining an advantage function of the agent by using the state value function, wherein the advantage function is used to evaluate the advantage of the first adjustment amount and the second adjustment amount in each interaction; Update the network parameters of the agent's policy network by the advantage function, and update the network parameters of the agent's value network by the estimated return and reward value; When the number of iterative training is greater than or equal to the number threshold, the iterative training is stopped, and the policy network and the value network after the last iterative training are determined as the target intelligent agent.

7. The method according to claim 6, characterized in that Updating the network parameters of the agent's strategy network by the advantage function, and updating the network parameters of the agent's value network by the estimated return and reward value include: Determine a trimming objective function of the policy network, input the advantage function into the trimming objective function, and update network parameters of the policy network until a function value of the trimming objective function reaches a maximum value; The estimated return and the reward value are input into a mean square error formula, and the network parameters of the value network are updated until the output value of the mean square error formula is the minimum value.

8. A device for configuring business data, characterized in that: include: An acquisition unit, configured to acquire data to be configured of a target institution, wherein the data to be configured includes at least one of the following: current business data, static business data, liquidity coverage ratio, and the rate of return of the static business data; An input unit, used to input the data to be configured into a target model to obtain a target adjustment amount of the data to be configured, wherein the target model simulates the environment of the target mechanism, controls the agent of the target model to interact with the environment based on the adjustment amount, and outputs the target adjustment amount that maximizes the benefit; The adjustment unit is used to adjust the data to be configured according to the target adjustment amount to obtain the target data after the target mechanism is adjusted.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the computer-readable storage medium is located is controlled to execute the business data configuration method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program executes the business data configuration method described in any one of claims 1 to 7 when running.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the business data configuration method described in any one of claims 1 to 7 are implemented.

Images