JP7669693B2 - Device for learning and predicting activation energy in chemical reactions - Google Patents

Description

The present invention relates to a learning device, a prediction device, a learning program, a prediction program, a learning method, and a prediction method for activation energy in chemical reactions.

Activation energy is the energy required for a chemical reaction of reactants to proceed, and is an important factor for estimating the ease with which a chemical reaction will proceed in various fields such as the chemical industry and pharmaceuticals. Therefore, there is a demand for a method to calculate the magnitude of activation energy in a realistic calculation time and environment.

Here, a model-based prediction technology has been proposed to predict the excited states of reactants, new materials, new combinations, etc.

For example, a method has been proposed in which machine learning is performed using data on the ground and excited states of molecules, and the properties of the excited states of molecules are predicted using a model generated by the machine learning (see, for example, Non-Patent Document 1).

Kawai et al. "Predicting excited states from ground state wavefunction by supervised quantum machine learning", Mach. Learn.: Sci. Technol. 1 (2020) 045027

When trying to directly calculate activation energy using quantum chemical calculations, there is a problem in that it is difficult to search for transition states. In addition, because the calculation method used to calculate activation energy is different from the calculation method used to calculate excited states, there is a problem in that the technology in Non-Patent Document 1 cannot be applied.

One aspect of the present invention aims to provide a learning device that can learn to predict activation energy with high accuracy.

One aspect of the learning device according to the present invention includes a first acquisition unit that acquires a first explanatory variable including structural parameters and electronic properties of a reaction system calculated by quantum chemical calculation, a second acquisition unit that acquires a second explanatory variable including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculation, a third acquisition unit that acquires a target variable including an activation energy in a chemical reaction of the reaction system calculated by quantum chemical calculation, and a learning unit that learns a model using learning data in which the first explanatory variable, the second explanatory variable, and the target variable are associated.

One aspect of the prediction device according to the present invention includes a first acquisition unit that acquires a first explanatory variable including structural parameters and electronic properties of a reaction system calculated by quantum chemical calculations, a second acquisition unit that acquires a second explanatory variable including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculations, a trained model trained using training data in which a first explanatory variable and a second explanatory variable calculated in advance for a specific reaction system are associated with an objective variable including an activation energy in a chemical reaction of the specific reaction system, and a prediction unit that outputs an objective variable including an activation energy in a chemical reaction of the reaction system predicted by the trained model by inputting the first explanatory variable and the second explanatory variable of the reaction system to be predicted.

One aspect of the learning program of the present invention causes a computer to execute a first acquisition step of acquiring first explanatory variables including structural parameters and electronic properties of a reaction system calculated by quantum chemical calculations, a second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculations, a third acquisition step of acquiring a target variable including an activation energy in a chemical reaction of the reaction system calculated by quantum chemical calculations, and a learning step of learning a model using learning data in which the first explanatory variables, the second explanatory variables, and the target variables are associated.

One aspect of the prediction program according to the present invention causes a computer to execute a first acquisition step of acquiring a first explanatory variable including a structural parameter and electronic property of a reaction system calculated by quantum chemical calculation, a second acquisition step of acquiring a second explanatory variable including a structural parameter and electronic property of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculation, and a prediction step of inputting the first explanatory variable and the second explanatory variable of a reaction system to be predicted into a trained model trained using training data in which a first explanatory variable and a second explanatory variable calculated in advance for a specific reaction system correspond to a target variable including an activation energy in a chemical reaction of the specific reaction system, thereby outputting a target variable including an activation energy in a chemical reaction of the reaction system to be predicted, which is predicted by the trained model.

One aspect of the learning method according to the present invention includes a first acquisition step of acquiring first explanatory variables including structural parameters and electronic properties of a reaction system calculated by quantum chemical calculations, a second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculations, a third acquisition step of acquiring a target variable including an activation energy in a chemical reaction of the reaction system calculated by quantum chemical calculations, and a learning step of learning a model using learning data in which the first explanatory variables, the second explanatory variables, and the target variables are associated.

One aspect of the prediction method according to the present invention includes a first acquisition step of acquiring a first explanatory variable including a structural parameter and electronic property of a reaction system calculated by quantum chemical calculation, a second acquisition step of acquiring a second explanatory variable including a structural parameter and electronic property of a product system generated by chemically reacting the reaction system calculated by quantum chemical calculation, and a prediction step of inputting the first explanatory variable and the second explanatory variable of a reaction system to be predicted into a trained model trained using training data in which a first explanatory variable and a second explanatory variable calculated in advance for a predetermined reaction system correspond to a target variable including an activation energy in a chemical reaction of the predetermined reaction system, thereby outputting a target variable including an activation energy in a chemical reaction of the reaction system to be predicted, which is predicted by the trained model.

One aspect of the learning device of the present invention can train a system to predict activation energy with high accuracy.

FIG. 1 is a diagram showing the change in free energy as the reaction progresses from a reaction system to a product system. 2A and 2B are block diagrams showing the schematic configuration of a learning device and a prediction device according to an embodiment of the present invention, where FIG. 2A is a block diagram showing the schematic configuration of a learning device, and FIG. 2B is a block diagram showing the schematic configuration of a prediction device. FIG. 2 is a block diagram showing a schematic configuration of a learning unit. 4A and 4B are diagrams showing an example of learning data and prediction data, where FIG. 4A shows learning data and FIG. 4B shows prediction data. FIG. 2 is a block diagram showing a schematic configuration of a prediction unit. FIG. 2 is a block diagram showing the hardware configuration of a learning device and a prediction device. 1 is a flowchart illustrating a learning method according to an embodiment of the present invention. 1 is a flowchart illustrating a prediction method according to an embodiment of the present invention. FIG. 13 is a diagram showing the distribution of values of a response variable stored in a data table. FIG. 13 is a diagram showing prediction verification results of learning data in a machine learning model. FIG. 13 is a diagram in which calculated values and predicted values are plotted together with training data. FIG. 1 is a plot of calculation time versus the number of electrons in the system. FIG. 13 is a diagram showing the distribution of the number of electrons in the system for 94 learning data patterns and prediction targets A to E.

The following is a detailed description of an embodiment of the present invention. To facilitate understanding of the description, the same components in each drawing are given the same reference numerals, and duplicate descriptions are omitted. In addition, in this specification, a tilde "~" indicating a numerical range means that the numerical values before and after it are included as the lower and upper limits, unless otherwise specified.

The learning device and prediction device according to this embodiment will be described. FIG. 1 is a diagram showing the change in free energy as the reaction progresses from the reaction system to the product system. As shown in FIG. 1, the learning device according to this embodiment performs machine learning using a first explanatory variable of the reaction system in a stable state, a second explanatory variable of the product system in a stable state, and a target variable including an activation energy obtained from the difference between the free energy of the reaction system and the free energy of the transition state, to generate a learned model.

The prediction device according to this embodiment predicts a response variable including activation energy by inputting a first explanatory variable of a reaction system in a stable state and a second explanatory variable of a product system in a stable state into a trained model in which the first explanatory variable and the second explanatory variable correspond to the response variable.

<Learning device>
A learning device according to this embodiment will be described. FIG. 2 is a block diagram showing a schematic configuration of the learning device and the prediction device according to this embodiment. As shown in FIG. 2(a), the learning device 1 includes a first acquisition unit 11, a second acquisition unit 12, a third acquisition unit 13, and a learning unit 14. The learning device 1 may further include a data table creation unit 15 and a display unit 16. The learning device 1 learns the value of the activation energy that occurs when reactants such as compounds undergo a chemical reaction, pass through a transition state, and generate a product.

The chemical reactions to be studied are not particularly limited as long as they are chemical reactions in which a substance passes through a transition state to produce a product. For example, the Diels-Alder reaction, which is an organic chemical reaction in which an alkene is added to a conjugated diene to produce a six-membered ring structure, can be mentioned.

The first acquisition unit 11 acquires a first explanatory variable including structural parameters and electronic properties of the reaction system calculated by quantum chemical calculation.

The first explanatory variables are various pieces of information about the reaction system, such as structural parameters and electronic properties, obtained by low-precision quantum chemical calculations (low-precision quantum chemical calculations). Details of the first explanatory variables will be described later.

In this embodiment, low accuracy varies depending on the conditions of the reaction system, and means that the accuracy of various information calculated under high-accuracy calculation conditions is lower than that of the information calculated under high-accuracy calculation conditions. The meaning of low accuracy below is used in the same sense. The accuracy of the calculation conditions can be determined by the method and basis set used in quantum scientific calculations, as described below.

The second acquisition unit 12 acquires second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, which are calculated by quantum chemical calculations.

The second explanatory variables are various pieces of information about the product, such as structural parameters and electronic properties, obtained by low-precision quantum chemical calculations. Details of the second explanatory variables will be described later.

The third acquisition unit 13 acquires a target variable including the activation energy in the chemical reaction of the reaction system, calculated by quantum chemical calculation.

The objective variables are various pieces of information, including the activation energy, which can be obtained by highly accurate quantum chemical calculations (high-precision quantum chemical calculations). Details of the objective variables will be described later.

In this embodiment, high accuracy varies depending on the conditions of the reaction system, etc., and refers to the accuracy of various information calculated under high accuracy calculation conditions. The meaning of high accuracy below is also used in the same sense.

The first explanatory variable, the second explanatory variable, and the objective variable are all calculated by quantum chemical calculations as described above. Quantum chemical calculations can be performed using general-purpose software, and for example, Gaussian16 (manufactured by Gaussian Inc.) can be suitably used.

When using Gaussian 16 to find a stable molecular structure, you can input the calculation method, basis set, initial coordinates of the molecular model, charge, and spin multiplicity to create an input file and execute the calculation to find the stable molecular structure.

Quantum chemical calculation methods include the Hartree-Fock method and density functional theory (DFT) method, but the DFT method is preferred because it provides a practically satisfactory balance between costs, such as calculation time, and calculation accuracy.

It should be noted that in the DFT method, the calculation results can vary significantly depending on which functional is selected. Therefore, for calculations that require accuracy, it is preferable to select a functional that includes long-range correction (taking into account the effects acting between two particles at long distances) and dispersion force correction (taking into account weak intermolecular interactions such as van der Waals forces). Representative functionals that combine long-range correction and dispersion force correction include, for example, ωB97XD and M06-2X.

As an example of the hierarchy of some of the functionals that can be used in Gaussian16, for the M06 system it is M06-L, M06, M06-2X, and for the ωB97 system it is ωB97, ωB97X, ωB97XD, which increases the calculation accuracy but also increases the calculation cost.

As with quantum chemical calculation methods, the basis set is an important factor that affects the calculation accuracy and calculation cost. In general, in order to reproduce molecular orbitals more accurately, it is desirable to use a basis set that can be assigned as many functions as possible. However, the higher the level of accuracy, the higher the calculation cost. Therefore, it is necessary to select a basis set that is within the range that can be used in reality. For example, the 3-21G basis set can be calculated at a low calculation cost. However, the accuracy is low, and the quantitativeness of the results obtained is not guaranteed. On the other hand, the 6-311++G (3df, 3pd) basis set can obtain highly accurate results. However, the calculation cost is high, and in the case of systems with large molecular sizes, it may not be possible to obtain results within a realistic time frame.

As an example, the hierarchy of basis functions in the Pople system is as follows: STO-3G, 3-21G, 3-21G ^* , 6-31G, 6-31G(d), 6-311G(d), 6-311+G(d), 6-311+G(d,p), 6-311+G(2d,p), 6-311++G(2d,p), 6-311++G(3df,3pd), in that order, the calculation accuracy increases but the calculation cost increases.

The learning unit 14 performs machine learning of the model using learning data in which the first explanatory variable, the second explanatory variable, and the objective variable are associated.

Figure 3 is a block diagram showing a schematic configuration of the learning unit 14. As shown in Figure 3, the learning unit 14 has a readout unit 141, a model M1, and a prediction accuracy verification unit 142.

The reading unit 141 reads the learning data from the learning data table created by the data table creation unit 15.

Model M1 is a model that calculates a response variable including an activation energy value from a first explanatory variable and a second explanatory variable.

Model M1 uses the computing power of a computer to perform machine learning on target data and make predictions on unknown data. Generally, a programming language such as Python is often used for programs to execute machine learning, and a development environment with functions such as numerical calculation and data analysis must be created. To easily create a machine learning environment, it is preferable to use a method such as Anaconda (registered trademark), which is software distributed by Anaconda, Inc. in the United States.

Anaconda (registered trademark) includes libraries commonly used in Python and machine learning, and these can be installed all at once, making it easy to set up a machine learning environment. The libraries included in Anaconda (registered trademark) are also useful for data analysis before running machine learning, which will be described next.

The learning unit 14 can check the learning data. The learning unit 14 can check the distribution of values for each variable and the presence or absence of correlation between variables for the first explanatory variable, the second explanatory variable, and the objective variable for each sample stored in the learning data table created by the data table creation unit 15.

The distribution of values can be visually confirmed by plotting a histogram. Generally, the more diverse the data set to be trained is, the more likely it is that a wide range of unknown data can be predicted, so it is preferable that the variable distribution of the training data is as broad and unbalanced as possible. In particular, when the distribution of the objective variable is extremely biased, it is preferable to add sample data to the training data that takes values in a range that reduces the bias.

The presence or absence of correlation between variables can be estimated by calculating the correlation coefficient between two sets of variable data. The correlation coefficient can be calculated, for example, using the following formula (1).

The presence or absence of correlation between variables can also be visually confirmed by creating a scatter plot of two sets of variable data for all combinations of variables. When an explanatory variable exists that is strongly correlated with the objective variable, it is highly likely that the explanatory variable contributes greatly to determining the objective variable. On the other hand, an explanatory variable that shows no correlation at all with the objective variable is unrelated to the objective variable and is highly likely to be unnecessary for predicting unknown data. It is also generally known that when there is a strong correlation between two explanatory variables, multicollinearity can lead to a deterioration in prediction accuracy. Therefore, in this case, it is effective to take measures such as deleting one of the explanatory variables from the training data.

Model M1 can perform regression analysis on prepared learning data to calculate a response variable from a first explanatory variable and a second explanatory variable. There are generally various regression methods used when predicting a response variable that takes a continuous value, such as linear regression and random forest methods, but it is preferable to adopt a method that suits the target data set while looking at the verification results of the prediction accuracy after machine learning.

The prediction accuracy verification unit 142 verifies (checks) the prediction accuracy of the model M1 obtained by regression analysis. One method for verifying prediction accuracy is, for example, cross-validation. Cross-validation divides prepared data into several groups, one of which is used as test data and the rest as learning data. Then, learning and validation are repeated the same number of times as the number of divided groups while changing the learning data and test data, and the obtained results are averaged to evaluate the prediction accuracy of the model. When the amount of data is limited, leave-one-out cross-validation, in which only one piece of the entire data is used as test data, is particularly useful.

Accuracy evaluation indices that indicate the degree of agreement between the predicted values by model M1 and the actual measured values include ^R2 (coefficient of determination), RMSE (root mean square error), MAE (mean absolute error), etc. If it is determined from these indices that the prediction accuracy of model M1 is insufficient, the prediction accuracy verification unit 142 goes back to checking the data again and adds data to model M1 or makes corrections such as adding or reducing explanatory variables in order to reduce data bias.

The completed model M1 is saved so that it can be called up when predicting the target (unknown substance).

Various regression methods and algorithms for verifying prediction accuracy are available in Scikit-learn, a library included in the above-mentioned Anaconda (registered trademark).

As shown in FIG. 2(a), the data table creation unit 15 creates learning data using the first explanatory variable, the second explanatory variable, and the objective variable obtained by the first acquisition unit 11, the second acquisition unit 12, and the third acquisition unit 13. An example of the learning data is shown in FIG. 4(a). As shown in FIG. 4(a), the data table creation unit 15 creates learning data composed of the first explanatory variable, the second explanatory variable, and the objective variable.

As sample substances to be included in the training data, it is preferable to prepare at least several dozen types of substances with different types and positional relationships of substituents, assuming that the elementary reaction mechanisms are common. Basically, only relatively small molecular sizes that do not require high calculation costs are required. However, in reactions where bulkiness directly affects the activation energy, it may be necessary to include sample substances with large molecular sizes and bulky substituents in the training data.

The molecular structures of the reaction system (stable state before the reaction) and the product system (stable state after the reaction) of the sample substance are determined by quantum chemical calculations, and a first explanatory variable and a second explanatory variable are obtained as explanatory variables for learning.

Once structural optimization using quantum chemical calculations is complete, various information is output, including structural parameters such as bond lengths and bond angles, the potential energy of the system, orbital energy, dipole moment, charge distribution, etc.

The data table creation unit 15 extracts information about each sample substance and adds it to the learning data table.

The data table creation unit 15 determines the molecular structure of the reaction system and transition state of the sample substance by quantum chemical calculations, obtains learning objective variables including the activation energy value, and adds the learning data to the learning data table.

In this case, it is preferable that the calculation conditions be as accurate as possible. For transition state calculations, first, a suitable transition state structure is searched for in a sample with the simplest structure with the fewest substituents, and the obtained structure is used as a template to create input structures for other samples, thereby minimizing the number of trial and error steps. After the structure optimization calculation is completed, the activation energy is calculated from the difference in potential energy between the reaction system and the transition state.

The first explanatory variable and the second explanatory variable are preferably calculated by quantum chemical calculations with lower calculation accuracy than the objective variable. In other words, the quantum chemical calculations performed to obtain the first explanatory variable and the second explanatory variable need to have uniform calculation conditions, but do not need to be as highly accurate as when the objective variable of the reaction system is calculated by quantum chemical calculations, and lower accuracy is sufficient.

In this case, it is preferable that the learning data table be prepared in CSV format. Data in CSV format is versatile and can be handled by many spreadsheet programs.

As shown in FIG. 2, the display unit 16 can display information about the learning data used in learning the model M1 and information about the model M1.

Note that in this embodiment, the learning device 1 does not need to include the prediction accuracy verification unit 142.

<Prediction device>
A prediction device according to this embodiment will be described. Fig. 2(b) is a system configuration diagram showing the configuration of the prediction device according to this embodiment. As shown in Fig. 2(b), the prediction device 2 includes a first acquisition unit 21, a second acquisition unit 22, a trained model M2, and a prediction unit 23. The prediction device 2 may further include a prediction data table creation unit 24 and a display unit 25.

The first acquisition unit 21, like the first acquisition unit 11, acquires first explanatory variables including structural parameters and electronic properties of the reaction system calculated by quantum chemical calculations. The functions of the first acquisition unit 21 are similar to those of the first acquisition unit 11, so details are omitted.

The second acquisition unit 22, like the second acquisition unit 12, acquires second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting a reaction system, which are calculated by quantum chemical calculations. The functions of the second acquisition unit 22 are similar to those of the second acquisition unit 12, so details are omitted.

As shown in FIG. 1, the trained model M2 is trained using training data in which a first explanatory variable and a second explanatory variable calculated in advance for a specific reaction system are associated with a target variable including an activation energy in a chemical reaction of the specific reaction system. As the trained model M2, a model obtained by machine learning the model M1 provided in the learning device 1 described above can be used. Specifically, as the trained model M2, a machine learning model saved using Anaconda or the like can be used.

The prediction unit 23 inputs the first explanatory variable and the second explanatory variable of the reaction system to be predicted, and outputs a target variable including the activation energy of the chemical reaction of the reaction system to be predicted, which is predicted by the trained model M2.

Figure 5 is a block diagram showing a schematic configuration of the prediction unit 23. As shown in Figure 5, the prediction unit 23 has a readout unit 231.

The reading unit 231 reads the first explanatory variable and the second explanatory variable stored in the prediction data table.

The prediction unit 23 calls the trained model M2, reads the first explanatory variable and the second explanatory variable stored in the prediction data table by the reading unit 231 into the trained model M2, and outputs a predicted value of the objective variable including the activation energy of the chemical reaction.

As shown in FIG. 2(b), the prediction data table creation unit 24 creates a prediction data table using the first explanatory variable and the second explanatory variable acquired by the first acquisition unit 21 and the second acquisition unit 22. An example of the prediction data table is shown in FIG. 4(b). As shown in FIG. 4(b), the prediction data table stores the first explanatory variable and the second explanatory variable. The first explanatory variable and the second explanatory variable are similar to the first explanatory variable and the second explanatory variable used in the learning device 1 described above, and therefore details are omitted.

The prediction data table, like the learning data table, is preferably prepared in CSV format. Data in CSV format is versatile and can be handled by many spreadsheet programs.

The display unit 25 can display the objective variables including the activation energy of the chemical reaction of the reaction system to be predicted, which are predicted by the trained model M2.

The learning device 1 and the prediction device 2 shown in FIG. 2(a) and FIG. 2(b) may be configured as a learning system and a prediction system. That is, the learning device 1 is a single device such as a PC (Personal Computer) that includes each component within the device, but one or more of the components may be placed outside the device and connected via a network. For example, the learning data table may be provided on the cloud. In this case, the learning device 1 is configured as a learning system by the learning data table connected via a network. Similarly, the prediction device 2 shown in FIG. 2(b) may also be configured as a learning system by one or more of the components that are placed outside the device and connected via a network. For example, the prediction data table may be provided on the cloud. In this case, the prediction device 2 is configured as a prediction system by the prediction data table connected via a network.

<Hardware configuration of learning device 1 and prediction device 2>
Next, an example of the hardware configuration of the learning device 1 and the prediction device 2 will be described. FIG. 6 is a block diagram showing the hardware configuration of the learning device 1 and the prediction device 2. As shown in FIG. 6, the learning device 1 and the prediction device 2 are configured as an information processing device (computer), and can be physically configured as a computer system including a CPU (Central Processing Unit: processor) 101, which is an arithmetic processing unit, a RAM (Random Access Memory) 102 and a ROM (Read Only Memory) 103, which are main storage devices, an input device 104, an output device 105, a communication module 106, and an auxiliary storage device 107 such as a hard disk. These are connected to each other by a bus 108. The output device 105 and the auxiliary storage device 107 may be provided externally.

The CPU 101 controls the overall operation of the learning device 1 and the prediction device 2, and performs various information processing. The CPU 101 executes the learning program and the prediction program stored in the ROM 103 or the auxiliary storage device 107, and controls the display operation of the measurement recording screen and the analysis screen.

RAM 102 is used as a work area for CPU 101 and may include non-volatile RAM for storing main control parameters and information.

ROM 103 stores basic input/output programs and the like. The prediction program may be stored in ROM 103.

The input device 104 includes a keyboard, a mouse, operation buttons, a touch panel, etc.

The output device 105 is a monitor display or the like. The output device 105 displays prediction results and the like, and the screen is updated in response to input/output operations via the input device 104 and the communication module 106.

The communication module 106 is a data transmission/reception device such as a network card, and functions as a communication interface that receives information from an external data recording server, etc., and outputs analysis information to other electronic devices.

The auxiliary storage device 107 is a storage device such as an SSD (Solid State Drive) or an HDD (Hard Disk Drive), and stores, for example, various data, files, etc. required for the operation of the learning device 1 and the prediction device 2.

The functions of the learning device 1 and the prediction device 2 shown in FIG. 2 are realized by loading predetermined computer software (including a learning program and a prediction program) into the main storage device such as the CPU 101 and the RAM 102 or the auxiliary storage device 107, and executing the learning program and the prediction program stored in the RAM 102, the ROM 103 or the auxiliary storage device 107 by the CPU 101. The functions of the learning device 1 and the prediction device 2 are realized by operating the input device 104, the output device 105 and the communication module 106 and reading and writing data in the RAM 102, the ROM 103 and the auxiliary storage device 107. That is, by executing the learning program and the prediction program according to this embodiment on a computer, the learning device 1 and the prediction device 2 function as the components of FIG. 2.

The learning program and prediction program are stored, for example, in a storage device provided in the computer. Note that the learning program and prediction program may be configured so that a part or all of them are transmitted via a transmission medium such as a communication line, and are received and recorded (including installed) by a communication module 106 or the like provided in the computer. Also, the learning program and prediction program may be configured so that a part or all of them are recorded (including installed) in the computer from a state in which they are stored in a portable storage medium such as a CD-ROM, DVD-ROM, or flash memory.

<Study Program>
The learning program according to this embodiment can use a program having the following configuration.
That is, the learning program according to this embodiment includes a first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
A third acquisition step of acquiring a target variable including an activation energy in a chemical reaction of the reaction system, the activation energy being calculated by a quantum chemical calculation;
a learning process of learning a model using learning data in which the first explanatory variable, the second explanatory variable, and the objective variable are associated with each other;
A program for causing a computer to execute at least the above can be used.

<Prediction Program>
The prediction program according to this embodiment can use a program having the following configuration.
That is, the prediction program according to the present embodiment includes a first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
a prediction step of inputting the first explanatory variable and the second explanatory variable of a reaction system to be predicted into a trained model trained using learning data in which a first explanatory variable and a second explanatory variable calculated in advance for a predetermined reaction system correspond to a response variable including an activation energy in a chemical reaction of the predetermined reaction system, thereby outputting a response variable including an activation energy in a chemical reaction of the reaction system to be predicted, which is predicted by the trained model;
A program for causing a computer to execute at least the above can be used.

<Learning Method>
Next, a learning method according to the present embodiment will be described. The learning method according to the present embodiment is a method for learning a model M1 using learning data in which a first explanatory variable, a second explanatory variable, and a target variable are associated with each other in a learning device 1 having a configuration as shown in FIG.

Figure 7 is a flowchart explaining the learning method according to this embodiment. As shown in Figure 7, the first acquisition unit 11 acquires first explanatory variables including structural parameters and electronic properties of the reaction system calculated by quantum chemical calculations (first acquisition process: step S11).

Next, the second acquisition unit 12 acquires second explanatory variables including the structural parameters and electronic properties of the product system generated by chemically reacting the reaction system, which are calculated by quantum chemical calculations (second acquisition process: step S12).

Next, the third acquisition unit 13 acquires a target variable including the activation energy in the chemical reaction of the reaction system, calculated by quantum chemical calculation (third acquisition process: step S13).

Next, the data table creation unit 15 creates a learning data table including the first explanatory variable, the second explanatory variable, and the objective variable (learning data table creation process: step S14). The data table creation unit 15 links the input first explanatory variable, second explanatory variable, and objective variable to create a learning data table such as that shown in FIG. 4(a), for example.

Next, the learning unit 14 learns the model M1 using the learning data in which the first explanatory variable and the second explanatory variable are associated with the objective variable (learning process: step S15). The learning unit 14 learns the model M1 in response to the input of the first explanatory variable and the second explanatory variable included in the learning data, so that the output matches the objective variable linked to these explanatory variables.

Next, the display unit 16 displays information about the learning data used in learning the model M1 and information about the model M1 (display process: step S16).

The first explanatory variable acquisition process (step S11) may be performed simultaneously with the second explanatory variable acquisition process (step S12), or may be performed after the second explanatory variable acquisition process (step S12).

<Prediction method>
Next, a prediction method according to the present embodiment will be described. The prediction method according to the present embodiment is a method for outputting a response variable including an activation energy in a chemical reaction of a reaction system to be predicted, which is predicted by a trained model M2 in a prediction device 2 having a configuration as shown in FIG.

Figure 8 is a flowchart explaining the prediction method according to this embodiment. As shown in Figure 8, the first acquisition unit 21 acquires first explanatory variables including structural parameters and electronic properties of the reaction system calculated by quantum chemical calculations (first acquisition process: step S21).

Next, the second acquisition unit 22 acquires second explanatory variables including the structural parameters and electronic properties of the product system generated by chemically reacting the reaction system, which are calculated by quantum chemical calculations (second acquisition process: step S22).

Next, the prediction data table creating unit 24 creates a prediction data table including the first explanatory variable and the second explanatory variable (prediction data table creating process: step S23). The prediction data table creating unit 24 creates a prediction data table including the input first explanatory variable and second explanatory variable, as shown in FIG. 3.

Next, the prediction unit 23 inputs the first explanatory variable and the second explanatory variable of the reaction system to be predicted into the trained model M2, and outputs a target variable including the activation energy in the chemical reaction of the reaction system to be predicted, which is predicted by the trained model M2 (prediction process: step S24). The prediction unit 23 inputs the first explanatory variable and the second explanatory variable input into the prediction data table in the prediction data table creation process (step S23) into the trained model M2, and outputs the target variable including the activation energy output from the trained model M2.

Next, the display unit 25 displays information on the objective variables, including the activation energy, output by the prediction unit 23 (display process: step S25).

The first explanatory variable acquisition process (step S21) may be performed simultaneously with the second explanatory variable acquisition process (step S22), or may be performed after the second explanatory variable acquisition process (step S22).

The effects of this embodiment will be described. The learning device 1 according to this embodiment includes a learning unit 14, and can perform machine learning of the model M1 to calculate a value that is close to the objective variable calculated when a high-precision quantum chemical calculation is performed from the first explanatory variable and the second explanatory variable. When determining the objective variable from the first explanatory variable and the second explanatory variable, generally, if a high-precision calculation condition is selected to perform a high-precision quantum chemical calculation, the objective variable can be calculated with high precision, but the burden of the calculation is large and a high calculation cost is required to obtain the calculation result. The learning device 1 can learn to predict the activation energy of a chemical reaction with high precision from the input first explanatory variable and second explanatory variable easily and at low calculation cost.

The learning device 1 can also reduce the burden and time required for calculations. Therefore, the learning device 1 can learn to predict the activation energy of a chemical reaction from the input first explanatory variable and second explanatory variable easily and with low calculation cost.

In addition, in the learning device 1, the first explanatory variable and the second explanatory variable can be calculated by quantum chemical calculations that have lower calculation accuracy than the objective variable. This allows the prediction device 2 to reduce the calculation load required to predict the objective variable.

The learning device 1 can also include a display unit 16. This allows the prediction device 2 to visually present to the user information for generating learning data and information about the model M1, allowing the user to easily grasp the learning data and information about the model M1.

The prediction device 2 according to this embodiment includes a trained model M2 and a prediction unit 23. As a result, when predicting a response variable from a first explanatory variable and a second explanatory variable, the prediction device 2 can predict an approximate or equivalent response variable when calculated from the first explanatory variable and the second explanatory variable by high-precision quantum chemical calculation. Therefore, the prediction device 2 can predict the activation energy of a chemical reaction with high precision.

In addition, by being equipped with the trained model M2 and the prediction unit 23, the prediction device 2 can predict the objective variable from the first explanatory variable and the second explanatory variable more easily than when a high-precision quantum chemical calculation is performed using a high-precision computer, and can reduce the burden and time required to calculate the objective variable. Therefore, the prediction device 2 can predict the activation energy of a chemical reaction simply and with low calculation cost.

Here, activation energy is the energy required for a chemical reaction to proceed, and is an important factor for estimating the ease with which a chemical reaction will proceed in various fields such as the chemical industry and pharmaceuticals. In general, there are experimental methods, computational science methods, etc., for determining the activation energy of a chemical reaction. In the experimental method, for example, an experiment is carried out multiple times at different temperatures to determine the rate constant by measuring the change in concentration of a substance during the reaction, and the activation energy can be determined from the slope of the straight line in the Arrhenius plot. However, this requires a great deal of effort, and in some reactions a straight line cannot be obtained, making it impossible to determine the activation energy. Therefore, it is difficult to accurately and reliably determine the activation energy using the experimental method.

In the computational science method, the activation energy can be directly measured from the difference between the free energy of the starting material in a stable state and the free energy of the transition state, as shown in FIG. 1, by determining the structure of the transition state of the reaction using quantum chemical calculations. However, when determining the activation energy directly using quantum chemical calculations, it is difficult to specify the structure of the transition state, and it is not easy to calculate the transition state in quantum chemical calculations, so it is difficult to predict the activation energy. Therefore, when searching for the structure of the transition state using quantum chemical calculations, the molecular structure of the substance in the transition state is estimated in advance. However, if the molecular structure of the substance cannot be estimated or the estimated molecular structure is far from the true transition state structure, the activation energy cannot be predicted correctly. When searching for the structure of the transition state using quantum chemical calculations, the value of the activation energy calculated by the engineer is likely to vary if it is performed based on the intuition or experience of the engineer. In some cases, it may take a lot of time and effort to trial and error. In addition, when calculating the activation energy by performing high-precision calculations using a highly reliable and high-precision computer, the burden and time required for the calculation increases. In particular, the larger the molecular size of the reaction system or product system, the more time and high-spec computers are required for calculations, making it difficult to balance time and calculation costs. Therefore, if calculation accuracy is sacrificed in order to perform calculations in a realistic calculation time and environment, only unreliable calculation results can be obtained, and as a result, the molecules that can be handled are effectively limited to small sizes.

The prediction device 2 uses various information on the reaction system and the product system obtained by low-precision quantum chemical calculation as the first explanatory variable and the second explanatory variable. The various information on the reaction system and the product system obtained by low-precision quantum chemical calculation is information on the stable structure of the reaction system and the product system, and can be easily and stably calculated by low-precision quantum chemical calculation. As shown in FIG. 1, the prediction device 2 can predict an approximate or equivalent activation energy ΔG when a transition state is obtained by high-precision quantum chemical calculation by applying the first explanatory variable and the second explanatory variable obtained by low-precision quantum chemical calculation to the trained model M2. Furthermore, the prediction device 2 can reduce the calculation time and calculation load compared to conventional high-precision quantum chemical calculation.

The prediction device 2 can therefore predict activation energy with high accuracy, simply and at low calculation cost, without being restricted by molecular size, by combining low-precision quantum chemical calculations and machine learning, and can therefore predict the objective variable with high accuracy even for reaction systems with relatively large molecular sizes. Therefore, the prediction device 2 can be suitably used to speed up research and development, for example, because it can quickly and accurately predict the activation energy of reaction systems with relatively large molecular sizes, such as materials and catalysts in the chemical industry.

The learning program according to this embodiment includes a learning process, and can cause a computer to execute machine learning of model M1. Therefore, by using the learning program according to this embodiment, it is possible to learn to predict the activation energy in a chemical reaction with high accuracy based on the input first explanatory variable and second explanatory variable.

In addition, the learning program according to this embodiment uses a first explanatory variable including the structural parameters and electronic properties of the reaction system and a second explanatory variable including the structural parameters and electronic properties of the product system, making it possible to predict the objective variable more easily than when using high-precision quantum chemical calculations, and to reduce the burden and time required for calculation. Therefore, the learning program according to this embodiment can learn to predict the activation energy of a chemical reaction simply and with low calculation cost from the input first explanatory variable and second explanatory variable.

The prediction program according to this embodiment includes a prediction step, and can be executed by a computer to output a target variable to be predicted. Therefore, when predicting a target variable from a first explanatory variable and a second explanatory variable, the prediction program according to this embodiment can predict an approximate or equivalent target variable when calculated from the first explanatory variable and the second explanatory variable by high-precision quantum chemical calculation. Therefore, the prediction device 2 can predict the activation energy of a chemical reaction with high precision.

In addition, the prediction program according to this embodiment includes a prediction process, and therefore can output the objective variable from the first explanatory variables and the second explanatory variables easily and quickly without selecting highly accurate calculation conditions and performing highly accurate quantum chemical calculations, and without increasing the burden and time required for calculation. This makes it possible to cause a computer to predict the objective variable, including the activation energy of a chemical reaction, simply and with low calculation cost.

The learning method according to this embodiment includes a learning process (step S13), and thus can perform machine learning of the model M1. Therefore, the learning method according to this embodiment can learn to predict the activation energy in a chemical reaction with high accuracy based on the input first explanatory variable and second explanatory variable.

In addition, the learning method according to this embodiment includes a learning process (step S13), which makes it possible to predict the objective variable from the first explanatory variable and the second explanatory variable more easily than when using high-precision quantum chemical calculations, and also reduces the burden and time required for calculation. Therefore, the learning method according to this embodiment can learn to predict the activation energy of a chemical reaction from the input first explanatory variable and second explanatory variable easily and at low calculation cost.

The prediction method according to this embodiment includes a prediction step (step S23), and can output a predicted value of the objective variable to be predicted. Therefore, when predicting an objective variable from a first explanatory variable and a second explanatory variable, the prediction method according to this embodiment can predict an approximate or equivalent objective variable when calculated from the first explanatory variable and the second explanatory variable by high-precision quantum chemical calculation. Therefore, the prediction method according to this embodiment can predict the activation energy of a chemical reaction with high precision.

In addition, the prediction method according to this embodiment includes a prediction step (step S23), and therefore can easily and quickly output the objective variable from the first explanatory variables and the second explanatory variables without selecting highly accurate calculation conditions and performing highly accurate quantum chemical calculations, and without increasing the burden and time required for calculating the objective variable. Therefore, the objective variable including the activation energy of a chemical reaction can be predicted easily and with high accuracy at low calculation cost.

The following examples are provided to further explain the embodiments, but the embodiments are not limited to these examples and comparative examples.

Example 1
[Machine learning environment]
In this embodiment, Anaconda (registered trademark) was used for machine learning, numerical calculation, and data analysis.

[Target reaction]
In this example, we will deal with the Diels-Alder reaction, which is an organic chemical reaction in which an alkene is added to a conjugated diene to produce a six-membered ring structure. The general formula for the Diels-Alder reaction is shown in the following formula (I). For simplicity, in this example, we will only consider the case where acyclic compounds react with each other.

（式中、ＴＳは遷移状態を表わし、Ｘ₁～Ｘ₄及びＹ₁～Ｙ₄は、それぞれ独立に、－Ｈ、－ＮＯ₂、－ＣＮ、－ＣＦ₃、－ＳＯ₂ＯＨ、－ＳＯ₂ＣＨ₃、－Ｐｈ、－ＣＨＯ、－ＣＯＯＨ、－ＣＯＯＣＨ₃、－Ｆ、－Ｃｌ、－Ｂｒ、－Ｃ（ＣＨ₃）₃、－ＣＨ₃、－ＯＣＨ₃及び－ＯＨ等の置換基の何れかを表わす。）

(In the formula, TS represents a transition state, and X ₁ to X ₄ and Y ₁ to Y ₄ each independently represent any one of substituents such as -H, -NO ₂ , -CN, -CF ₃ , -SO ₂ OH, -SO ₂ CH ₃ , -Ph, -CHO, -COOH, -COOCH ₃ , -F, -Cl, -Br, -C(CH ₃ ) ₃ , -CH ₃ , -OCH ₃ , and -OH.)

[Create model]
(Acquiring training data and creating a training data table)
In this embodiment, 94 patterns, No. 1 to No. 94, shown in Tables 1 to 3, were used to obtain learning data from information obtained from quantum chemical calculations for sample compounds with different combinations of substituents X ₁ to X ₄ and Y _{1 to} Y 4. In order to provide a distribution to the activation energy value, which is the objective variable, compounds combining 17 types of electron-donating and electron-withdrawing substituents (-H, -NO ₂ , -CN, -CF ₃ , -SO ₂ OH, -SO ₂ CH ₃ , -Ph, -CHO, -COOH, -COOCH ₃ , -F, -Cl, -Br, -C(CH ₃ ) ₃ , -CH ₃ , -OCH ₃ , -OH) were targeted. In order to reduce calculation costs and human labor, substituents with large molecular size and large degree of orientation freedom were avoided as much as possible.

For the 94 patterns No. 1 to No. 94 shown in Tables 1 to 3, a structural optimization calculation was performed using the Gaussian16 program by the DFT method. The calculation conditions and calculation targets for the first explanatory variable and the second explanatory variable and for the objective variable are shown in Table 4. The low-precision basis 3-21G ^* was used to obtain the first explanatory variable and the second explanatory variable, and the high-precision basis 6-311+G(2d,p) was used to obtain the objective variable data.

From the output calculation results, data for the first explanatory variable, the second explanatory variable, and the objective variable were extracted. The objective variable, activation energy, was calculated according to the following formula (2).
Activation energy = free energy of the transition state - (free energy of reactant 1 + free energy of reactant 2) ... (2)

As the first and second explanatory variables, data on a total of 54 types shown in Table 5 (27 variables derived from reactants 1 and 2, 26 variables derived from the product, and one variable for the stabilization energy of the product system) were extracted, and a learning data table was created along with the objective variable.

The carbon atom labels C1 to C6 are defined in accordance with reaction formula (II) below.

[Check the learning data]
The distribution of the objective variable values stored in the created learning data table is shown in Figure 9. As shown in Figure 9, the range of activation energy is from 27 kcal/mol to 43 kcal/mol, and since there is continuous data within this range, it can be said that there is generally no problem.

Tables 6 and 7 also show the correlation coefficients between the first and second explanatory variables and the objective variable, in descending order of absolute value of the correlation coefficient.

From Tables 6 and 7, it was confirmed that the variables derived from reactant 2 (alkene), the products, and the stabilization energy of the product system are explanatory variables that are relatively strongly correlated with the objective variable.

(Creating a learning model and verifying prediction accuracy)
Regression analysis and prediction accuracy verification were performed on the prepared learning data. Bayesian ridge regression was used as the regression method, and leave-one-out cross validation was used as the accuracy verification method. In addition, when the 54 explanatory variables with the weakest correlation with the objective variable were removed from the bottom, the model with the highest prediction accuracy (evaluated by ^R2 value) was adopted, using the top 52 correlated variables (i.e., excluding the bottom two correlated variables).

The results of predicting and verifying the learning data in the created learning model are shown in Figure 10. As shown in Figure 10, although there is some variation depending on the sample, the accuracy ^R2 value of the learning model is 0.834, and the MAE (mean absolute error) is 1.1 kcal/mol. It was confirmed that the activation energy can be obtained with almost the same accuracy for the learning data in Example 1 as when the activation energy is directly obtained by high-precision quantum chemical calculation.

[Prediction of activation energy of the predicted target (unknown substance)]
Using the completed trained model, the activation energy was predicted when an unknown substance not included in the training data, which is the prediction target, reacts. The prediction results are shown in Table 8. Table 8 shows five patterns of prediction targets A to E that were treated as _prediction targets. In Table 8, 1-Ad represents a 1-adamantyl group ( _-C10H15 ), and TBS represents a t-butyldimethylsilyl group (-Si( _CH3 ) _2C ( _CH3 ) ₃ ).

Low-precision quantum chemical calculations were performed for five prediction targets A to E. The calculation conditions and calculation targets are shown in Table 9. The first and second explanatory variables were extracted from the output calculation results, and a prediction data table was created.

The prediction data table we created was used as input, and the trained model was used to output the predicted activation energy as the objective variable. The output results are shown in Table 10.

<Verification of prediction results for the prediction target>
The above prediction results were compared with the results of the activation energy actually calculated by high-precision quantum chemical calculations to verify the prediction accuracy of the predicted value of the activation energy predicted using the trained model. The calculation conditions and calculation target for the verification are shown in Table 11.

The predicted values of the activation energies and the calculated values of the activation energies obtained are shown in Table 12. Figure 11 shows a plot of the predicted values and the calculated values together with the learning data.

As shown in Table 12 and Figure 11, it was confirmed that the predicted value of the activation energy predicted using the trained model created in this example was roughly the same as the calculated value obtained by high-precision quantum chemical calculation, even for untrained data. Therefore, it can be said that even for the unknown substance to be predicted, if the first explanatory variable and the second explanatory variable obtained by low-precision quantum chemical calculation of untrained data are applied to the trained model, it is possible to predict an activation energy equivalent to or close to the activation energy obtained by high-precision quantum chemical calculation.

In addition, the calculation cost, which is an issue when directly calculating activation energy by quantum chemical calculation, was compared with the prediction device method of this embodiment. In general, the calculation time depends on the number of atoms and electrons constituting the target molecule, or more precisely, the number of basis functions expressing each electronic state. Table 13 shows the number of atoms, number of electrons, number of basis functions, and calculation time (actual results) for the five prediction targets A to E for which data was obtained in this embodiment. FIG. 12 shows a plot of the calculation time against the number of electrons in the system. The required time is the sum of the time required to optimize the structures of reactant 1, reactant 2, and product, and the time required for vibration calculation. The number of computer cores was 16 in both cases.

As Comparative Example 1, a conventional method was used to directly calculate the activation energy by calculating the free energy of the reaction system and the free energy of the transition state using high-precision quantum chemical calculations. In Comparative Example 1, the required time was the sum of the time required to optimize the structures of reactant 1, reactant 2, and the transition state, and the time required for vibration calculation.

As shown in Table 13 and FIG. 12, the time required for calculation in Example 1 was shorter than that in Comparative Example 1. In particular, the effect of shortening the time required for calculation was greater as the molecular size increased. Therefore, it was confirmed that the conventional method of directly calculating activation energy using high-precision quantum chemical calculations requires more calculation time as the molecular size of the system being handled increases. In contrast, it was confirmed that the prediction device according to this embodiment can significantly shorten the calculation time because it is sufficient to apply the first explanatory variable and the second explanatory variable calculated using low-precision calculations to the trained model. It was also confirmed that the calculation time is prevented from increasing even when the molecular size increases, and a significant effect of shortening the calculation time is obtained.

Figure 13 shows the distribution of the number of electrons in the reaction system for 94 patterns of training data and five prediction targets A to E. As shown in Figure 13, the number of molecules in the reaction system was small in the training data, whereas the number of molecules in the reaction system was mostly large in the prediction targets A to E. In other words, it can be seen that the data group acquired when creating the trained model is contained in a relatively small molecular size region, whereas the prediction targets A to E are distributed in a region with larger molecular sizes than the trained data. Therefore, it can be said that the trained model created in this example can accurately predict the activation energy of prediction targets that exist in the untrained region with molecular sizes larger than the training data.

Therefore, by using the trained model created in this example, it is possible to accurately predict the activation energy of a reaction system that exists in a region of large molecular size, where high-precision quantum chemical calculations are difficult, by using small-sized samples that are relatively inexpensive and time-saving to obtain training data.

Although the embodiments have been described above, they are presented as examples, and the present invention is not limited to the above embodiments. The above embodiments can be implemented in various other forms, and various combinations, omissions, substitutions, modifications, etc. can be made without departing from the gist of the invention. These embodiments and their variations are included in the scope and gist of the invention, and are included in the scope of the invention and its equivalents described in the claims.

REFERENCE SIGNS LIST 1 Learning device 2 Prediction device 11, 21 First acquisition unit 12, 22 Second acquisition unit 13 Third acquisition unit 14 Learning unit 142 Prediction accuracy verification unit 15 Data table creation unit 16, 25 Display unit 23 Prediction unit 24 Prediction data table creation unit M1 Model M2 Learned model

Claims (9)

A first acquisition unit that acquires a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
a second acquisition unit that acquires second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
A third acquisition unit that acquires a target variable including an activation energy in a chemical reaction of the reaction system, the activation energy being calculated by a quantum chemical calculation;
a learning unit that performs model learning using learning data in which the first explanatory variable, the second explanatory variable, and the objective variable are associated with each other;
A learning device having the above configuration. The learning device according to claim 1, wherein the first explanatory variable and the second explanatory variable are calculated by a quantum chemical calculation having a lower calculation accuracy than the objective variable. The learning device according to claim 1 or 2, further comprising a display unit that displays information for generating learning data used in learning the model and information about the model to be learned. A first acquisition unit that acquires a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
a second acquisition unit that acquires second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
a trained model trained using training data in which a first explanatory variable and a second explanatory variable calculated in advance for a predetermined reaction system are associated with a target variable including an activation energy in a chemical reaction of the predetermined reaction system;
a prediction unit that receives a first explanatory variable and a second explanatory variable of a reaction system to be predicted, and outputs a target variable including an activation energy in a chemical reaction of the reaction system to be predicted, predicted by the trained model;
A prediction device having the above configuration. A reception unit that receives a selection of information regarding the reaction system of the prediction target input to the trained model and information regarding a product system generated by chemically reacting the reaction system of the prediction target;
a display unit that displays a response variable including an activation energy in a chemical reaction of the reaction system to be predicted;
The prediction device according to claim 4 , further comprising: A first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
A third acquisition step of acquiring a target variable including an activation energy in a chemical reaction of the reaction system, the activation energy being calculated by a quantum chemical calculation;
a learning process of learning a model using learning data in which the first explanatory variable, the second explanatory variable, and the objective variable are associated with each other;
A learning program that causes a computer to execute the following: A first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
a prediction step of inputting the first explanatory variable and the second explanatory variable of a reaction system to be predicted into a trained model trained using learning data in which a first explanatory variable and a second explanatory variable calculated in advance for a predetermined reaction system correspond to a response variable including an activation energy in a chemical reaction of the predetermined reaction system, thereby outputting a response variable including an activation energy in a chemical reaction of the reaction system to be predicted, which is predicted by the trained model;
A prediction program that causes a computer to execute the above. A first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
A third acquisition step of acquiring a target variable including an activation energy in a chemical reaction of the reaction system, the activation energy being calculated by a quantum chemical calculation;
a learning process of learning a model using learning data in which the first explanatory variable, the second explanatory variable, and the objective variable are associated with each other;
The computer executes the learning method. A first acquisition step of acquiring a first explanatory variable including a structural parameter and an electronic property of a reaction system calculated by a quantum chemical calculation;
A second acquisition step of acquiring second explanatory variables including structural parameters and electronic properties of a product system generated by chemically reacting the reaction system, the second explanatory variables being calculated by quantum chemical calculation;
a prediction step of inputting the first explanatory variable and the second explanatory variable of a reaction system to be predicted into a trained model trained using learning data in which a first explanatory variable and a second explanatory variable calculated in advance for a predetermined reaction system correspond to a response variable including an activation energy in a chemical reaction of the predetermined reaction system, thereby outputting a response variable including an activation energy in a chemical reaction of the reaction system to be predicted, which is predicted by the trained model;
A prediction method implemented by a computer .

Images