Information processing device and information processing program
An information processing device comprises a representation learning unit for learning characteristic vectors representing the various characteristics of objects, on the basis of a game log, which is game progress history related to an electronic game in which a plurality of objects are used and which comprises information about the game situation including information about objects that affect the game, information about objects used in said situation from among said objects, and information indicating the effect on the game arising from the use of said objects, wherein the characteristic vectors are found by performing learning using a combination of information about the effect on the game and information obtained by excluding the information about at least one of the objects from the information about the game situation.
1. An information processing device, comprising:
at least one hardware processor; and
one or more software modules configured to, when executed by the at least one hardware processor,
access a game log comprising a game progress history related to an electronic game in which a plurality of objects are used, wherein the game progress history represents each of a plurality of game situations, and one or more of the plurality of objects used in each of the plurality of game situations,
for each of the plurality of objects, derive a characteristic vector representing the object by, for each of one or more of the plurality of game situations in which the object is used, excluding the object as it was used in that same situation from that game situation to identify an effect of the object on that game situation, and
use the characteristic vectors for the plurality of objects in a machine learning process, to train an artificial intelligence (AI) agent to play the electronic game.
2. The information processing device according to claim 1 , wherein the one or more game situations from which each object is excluded comprise game situations in which that object was used during an attack.
3. The information processing device according to claim 1 , wherein the one or more game situations from which each object is excluded comprise game situations in which that object was used during a defense.
4. The information processing device according to claim 1 , wherein deriving the characteristic vectors comprises, for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
5. An information processing device according to claim 2 , wherein deriving the characteristic vectors comprises for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
6. An information processing device according to claim 3 , wherein deriving the characteristic vectors comprises, for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
7. The information processing device according to claim 1 , wherein the AI agent is trained to play the electronic game according to a strategy related to use of the plurality of objects.
8. A non-transitory computer-readable medium including instructions, wherein the instructions, when executed by a processor, cause the processor to:
access a game log comprising a game progress history related to an electronic game in which a plurality of objects are used, wherein the game progress history represents each of a plurality of game situation, and one or more of the plurality of objects used in each of the plurality of game situations;
for each of the plurality of objects, derive a characteristic vector representing the object by, for each of one or more of the plurality of game situations in which the object is used, excluding the object as it was used in that game situation from that game situation to identify an effect of the object on that game situation, and
use the characteristic vectors for the plurality of objects in a machine learning process, to train an artificial intelligence (AI) agent to play the electronic game.
9. The non-transitory computer-readable medium according to claim 8 , wherein the one or more game situations from which each object is excluded comprise game situations in which that object was used during an attack.
10. The non-transitory computer-readable medium according to claim 8 , wherein the one or more game situations from which each object is excluded comprise game situations in which that object was used during a defense.
11. The non-transitory computer-readable medium according to claim 8 , wherein deriving the characteristic vectors comprises, for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
12. The non-transitory computer-readable medium according to claim 9 , wherein deriving the characteristic vectors comprises, for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
13. The non-transitory computer-readable medium according to claim 10 , wherein deriving the characteristic vectors comprises, for each of the plurality of objects, combining a plurality of individual vectors obtained for that object into an overall characteristic vector for that object.
14. The non-transitory computer-readable medium according to claim 8 , wherein the AI agent is trained to play the electronic game according to a strategy related to use of the plurality of objects.