METHOD AND SYSTEM FOR ARTIFICIAL INTELLIGENCE BASED VIDEO GAME TESTING
A method for evaluating a build of a computer-implemented game is disclosed. An evaluation request is received. The evaluation. request includes an identification of the build, data describing one or more behaviors for the build, and data describing one or more tests. One or more simulations of playing of the build are performed using the one or more behaviors. One or more metrics ate extracted from the simulations. Each of the one or more metrics measures an aspect of the computer-implemented game. One or more tests are applied to the one or more metrics to evaluate an adherence of the build to the one or more tests. A display of the evaluation is caused to be displayed in a user interface of a client device.
1 . A system comprising:
one or more computer processors;
one or more computer memories;
a set of instructions incorporated into the one or more computer memories, the set of instructions configuring the one or more computer processors to perform operations for evaluating a build of a computer-implemented game, the operations comprising:
receiving an evaluation request, the evaluation request including an identification of the build, data describing one or more behaviors for the build, and data describing one or more tests;
performing one or more simulations of playing of the build using the one or more behaviors;
extracting one or more metrics from the simulations, each of the one or more metrics measuring an aspect of the computer-implemented game;
applying the one or more tests to the one or more metrics to evaluate an adherence of the build to the one or more tests; and
causing a display of the evaluation in a user interface of a client device.
2 . The system of claim 1 , the operations alternatively being for optimizing game parameters of the build of the computer-implemented game, the operations further comprising:
updating the parameters of the build based on execution of an optimization loop until each of the one or more tests is passed or until a cost budget is reached or exceeded; and
causing a display of results of the optimizing in the user interface of the client device.
3 . The system of claim 2 , wherein the execution of the optimization loop includes, at least, repeatedly performing the following operations:
modifying at least one parameter of the one or more game parameters within the build;
performing an additional one or more simulations of playing of the modified build using the one or more behaviors;
extracting an additional one or more metrics from the simulations, each of the additional one or more metrics measuring an aspect of the computer-implemented. game; and
applying the one or more tests to the additional one or more metrics to evaluate the adherence of the modified build to the one or more tests.
4 . The system of claim 3 , wherein a selection of the at least one parameter and a range for the modification of the at least one parameter is specified within the request.
5 . The system of claim 1 , wherein the behavior includes an artificial intelligence agent trained via a machine learning model to play the computer-implemented game.
6 . The system of claim 5 , wherein the model is parametrized to control the playing of the computer-implemented game and wherein the request may include a specification of the parametrization.
7 . The system of claim 1 , further including a testing server device, and wherein the operations further include:
sending the request over a network to the testing server device; and
receiving data pertaining to results of the simulations, the extracting of the values, and the applying of the tests from the testing server device, the simulations, the extracting of the values of the metrics, and the applying of the tests being performed on the server device.
8 . The system of claim 1 , wherein there are plurality of different builds of the computer-implemented game specified in the request, and wherein the metrics generated for each different build of the plurality of different builds are aggregated using a plurality of aggregation functions.
9 . The system of claim 1 , wherein the operations include:
providing a graphical user interface for display on a display device, the graphical user interface comprising a display area that includes tools for connecting to a behavior management service for generating, training and modifying a behavior.
10 . A non-transitory computer-readable storage medium comprising a set of instructions that, when executed by one or more computer processors, causes the one or more computer processors to perform operations for evaluating a build of a computer-implemented game, the operations comprising:
receiving an evaluation request, the evaluation request including an identification of the build, data describing one or more behaviors for the build, and data describing one or more tests;
performing one or more simulations of playing of the build using the one or more behaviors;
extracting one or more metrics from the simulations, each of the one or more metrics measuring an aspect of the computer-implemented game;
applying the one or more tests to the one or more metrics to evaluate an adherence of the build to the one or more tests; and
causing a display of the evaluation in a user interface of a client device.
11 . The non-transitory computer-readable medium of claim 10 , the operations alternatively being for optimizing game parameters of the build of the computer-implemented game, the operations further comprising:
updating the parameters of the build based on execution of an optimization loop until each of the one or more tests is passed or until a cost budget is reached or exceeded; and
causing a display of results of the optimizing in the user interface of the client device.
12 . The non-transitory computer-readable medium of claim 11 , wherein the execution of the optimization loop includes, at least, repeatedly performing the following operations:
modifying at least one parameter of the one or more game parameters within the build;
performing an additional one or more simulations of playing of the modified build using the one or more behaviors;
extracting an additional one or more metrics from the simulations, each of the additional one or more metrics measuring an aspect of the computer-implemented game; and
applying the one or more tests to the additional one or more metrics to evaluate the adherence of the modified build to the one or more tests.
13 . The non-transitory computer-readable medium of claim 12 , wherein a selection of the at least one parameter and a range for the modification of the at least one parameter is specified within the request.
14 . The non-transitory computer-readable medium of claim 10 , wherein the behavior includes an artificial intelligence agent trained via a machine learning model to play the computer-implemented game.
15 . The non-transitory computer-readable medium of claim 10 , wherein the model is parametrized to control the playing of the computer-implemented game and wherein the request may include a specification of the parametrization.
16 . The non-transitory computer-readable medium of claim 10 , further including a testing server device, and wherein the operations further include:
sending the request over a network to the testing server device; and
receiving data pertaining to results of the simulations, the extracting of the values, and the applying of the tests from the testing server device, the simulations, the extracting of the values of the metrics, and the applying of the tests being performed on the server device.
17 . The non-transitory computer-readable medium of claim 10 , wherein there are plurality of different builds of the computer-implemented game specified in the request, and wherein the metrics generated for each different build of the plurality of different builds are aggregated using a plurality of aggregation functions.
18 . The non-transitory computer-readable medium of claim 10 , wherein the operations include:
providing a graphical user interface for display on a display device, the graphical user interface comprising a display area that includes tools for connecting to a behavior management service for generating, training and modifying a behavior.
19 . A method comprising:
performing operations, using one or more computer processors, for evaluating a build of a computer-implemented game, the operations comprising:
receiving an evaluation request, the evaluation request including an identification of the build, data describing one or more behaviors for the build, and data describing one or more tests;
performing one or more simulations of playing of the build using the one or more behaviors;
extracting one or more metrics from the simulations, each of the one or more metrics measuring an aspect of the computer-implemented game;
applying the one or more tests to the one or more metrics to evaluate an adherence of the build to the one or more tests; and
causing a display of the evaluation in a user interface of a client device.
20 . The method of claim 19 , wherein the operations are alternatively for optimizing game parameters of the build of the computer-implemented game, the operations further comprising:
updating the parameters of the build based on execution of an optimization loop until each of the one or more tests is passed or until a cost budget is reached or exceeded, the execution of the optimization loop including repeatedly performing, at least, the following operations:
modifying at least one parameter of the one or more game parameters within the build;
performing an additional one or more simulations of playing of the modified build using the one or more behaviors;
extracting an additional one or more metrics from the simulations, each of the additional one or more metrics measuring an aspect of the computer-implemented game; and
applying the one or more tests to the additional one or more metrics to evaluate the adherence of the modified build to the one or more tests; and
causing a display of results of the optimizing in the user interface of the client device.