EP4100139A1 - Génération de contenu basée sur l'ia pour des applications de jeu - Google Patents

Génération de contenu basée sur l'ia pour des applications de jeu

Info

Publication number
EP4100139A1
EP4100139A1 EP21750106.3A EP21750106A EP4100139A1 EP 4100139 A1 EP4100139 A1 EP 4100139A1 EP 21750106 A EP21750106 A EP 21750106A EP 4100139 A1 EP4100139 A1 EP 4100139A1
Authority
EP
European Patent Office
Prior art keywords
game
game content
content
new
gaming
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP21750106.3A
Other languages
German (de)
English (en)
Inventor
Sam Snodgrass
Vanessa Volz
Niels Orsleff JUSTESEN
Sebastian RISI
Lars Henriksen
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Modl AI ApS
Original Assignee
Modl AI ApS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US17/159,907 external-priority patent/US11596867B2/en
Application filed by Modl AI ApS filed Critical Modl AI ApS
Publication of EP4100139A1 publication Critical patent/EP4100139A1/fr
Pending legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/60—Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor
    • A63F13/67—Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor adaptively or by learning from player actions, e.g. skill level adjustment or by storing successful combat sequences for re-use
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
    • A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
    • A63F13/60—Generating or modifying game content before or while executing the game program, e.g. authoring tools specially adapted for game development or game-integrated level editor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/047—Probabilistic or stochastic networks

Definitions

  • the present disclosure relates to processing systems and applications used in the development of gaming applications used by gaming systems and other gaming devices.
  • FIG. 1 presents a pictorial/block diagram representation of a game development system in accordance with an embodiment of the present disclosure.
  • FIG. 2 presents a block diagram representation of a game development platform in accordance with an embodiment of the present disclosure.
  • FIG. 3A presents a flow/block diagram representation of a game development pipeline in accordance with an embodiment of the present disclosure.
  • FIG. 3B presents a flow/block diagram representation of a components of the general experience personas in accordance with an embodiment of the present disclosure.
  • FIG. 4 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIG. 5 presents graphical representations of game telemetry data in accordance with an embodiment of the present disclosure.
  • FIG. 6 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIG. 7 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIG. 8 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIG. 9 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIGs. 10A and 10B present graphs in accordance with embodiments of the present disclosure.
  • FIG. IOC presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • FIG. 11 A presents a pictorial representation of an existing game in accordance with an embodiment of the present disclosure.
  • FIG. 1 IB presents a graphical representation of the existing game in accordance with an embodiment of the present disclosure.
  • FIG. llC presents a pictorial representation of an existing game in accordance with an embodiment of the present disclosure.
  • FIG. 1 ID presents a graphical representation of the existing game in accordance with an embodiment of the present disclosure.
  • FIG. 1 IE presents a table of accumulated configurations in accordance with the present disclosure.
  • FIG. 1 IF presents a table of conditional probabilities in accordance with the present disclosure.
  • FIG. 12A presents a pictorial representation of an existing game in accordance with an embodiment of the present disclosure.
  • FIG. 12B presents a pictorial representation of an existing game in accordance with an embodiment of the present disclosure.
  • FIG. 12C presents a graphical representation of an existing game in accordance with an embodiment of the present disclosure.
  • FIG. 12D presents a graphical representation of the existing game in accordance with an embodiment of the present disclosure.
  • FIG. 1 presents a pictorial/block diagram representation of a game development system in accordance with an embodiment of the present disclosure.
  • a game development platform 125 is presented that communicates game data 118 and player data 119 via network 115 with gaming devices such as mobile device 113 and gaming system 112 via network 115
  • the network 115 can be the Internet or other wide area or local area network.
  • the game development system 125 can be used in the creation, development, testing, balancing and updating of a gaming application.
  • the game data 118 can include, for example, a current version of a gaming application that is presented to the gaming devices for play. Furthermore, the game data 118 sent from the gaming devices to the game development platform 125 can include game telemetry data or be processed to produce game telemetry data and/or other game analytics used in game development.
  • the player data 119 can include one or more modes of output such as player or viewer verbal data generated by a microphone associated with the gaming system 112 or 113, chat data associated with a player or viewer and/or non-verbal data of a player or viewer such as facial expression, head pose, that is captured via a camera or other imaging sensor associated with the gaming system 112 or 113 that indicates, for example, player and/or viewer emotions.
  • FIG. 2 presents a block diagram representation of a game development platform in accordance with an embodiment of the present disclosure.
  • the game development platform 125 includes a network interface 220 such as a 3G, 4G, 5G or other cellular wireless transceiver, a Bluetooth transceiver, a WiFi transceiver, UltraWideBand transceiver, WIMAX transceiver, ZigBee transceiver or other wireless interface, a Universal Serial Bus (USB) interface, an IEEE 1394 Firewire interface, an Ethernet interface or other wired interface and/or other network card or modem for communicating for communicating with one or more gaming devices via network 115.
  • a network interface 220 such as a 3G, 4G, 5G or other cellular wireless transceiver, a Bluetooth transceiver, a WiFi transceiver, UltraWideBand transceiver, WIMAX transceiver, ZigBee transceiver or other wireless interface, a Universal Serial Bus (USB) interface, an IEEE 1394 Fire
  • the game development platform 125 also includes a processing module 230 and memory module 240 that stores an operating system (O/S) 244 such as an Apple, Unix, Linux or Microsoft operating system or other operating system, a game development application 246, one or more gaming applications 248, one or more gaming bots 250, one or more procedural content generation (PCG) tools 252, and one or more behavioral experience analysis (BEA) tools 254.
  • O/S operating system
  • game development application 246, gaming application 248, gaming bots 250, PCG tools 252 and BEA tools 254 each include operational instructions that, when executed by the processing module 230, that cooperate to configure the processing module into a special purpose device to perform the particular functions described herein.
  • the game development platform 125 also includes a user interface (I/F) 262 such as a display device, touch screen, key pad, touch pad, joy stick, thumb wheel, a mouse, one or more buttons, a speaker, a microphone, an accelerometer, gyroscope or other motion or position sensor, video camera or other interface devices that provide information to a user of the game development platform 125 and that generate data in response to the user’s interaction with the game development platform 125.
  • I/F user interface
  • the processing module 230 can be implemented via a single processing device or a plurality of processing devices.
  • processing devices can include a microprocessor, micro controller, digital signal processor, microcomputer, central processing unit, field programmable gate array, programmable logic device, state machine, logic circuitry, analog circuitry, digital circuitry, and/or any device that manipulates signals (analog and/or digital) based on operational instructions that are stored in a memory, such as memory 240.
  • the memory module 240 can include a hard disc drive or other disc drive, read-only memory, random access memory, volatile memory, non-volatile memory, static memory, dynamic memory, flash memory, cache memory, and/or any device that stores digital information.
  • the memory storing the corresponding operational instructions may be embedded within, or external to, the circuitry comprising the state machine, analog circuitry, digital circuitry, and/or logic circuitry. While a particular bus architecture is presented that includes a single bus 260, other architectures are possible including additional data buses and/or direct connectivity between one or more elements. Further, the game development platform 125 can include one or more additional elements that are not specifically shown.
  • the game development application 246 can be used by a game developer in the creation, development, testing, balancing, improving, revision, optimizing and/or updating of the gaming application 248.
  • the gaming application 248 can be, for example, a multiplayer or single player game including a shooter or other combat game, fantasy game or other action or adventure game, a simulation game that simulates the operation of a real-world vehicle device or system, a realtime strategy game, a puzzle, a sports game, role-playing game, board game or other video or digitally animated game.
  • one or more versions of the gaming application 248 can be stored including, for example, multiple versions or updates of the gaming application, one or more sets of game parameters, one or more levels and other content and/or other gaming data.
  • the gaming bots 250 operate in conjunction with the game development application 246 to test the operation of the gaming application 246 and/or to operate as one or more non player characters (NPCs) in the game.
  • the gaming bots 250 can include and/or operate as game playing AI (artificial intelligence) personas that are constructed and implemented via a machine learning algorithm and that operate, for example, as automatic testers designed to represent specific play-styles or skill levels.
  • AI artificial intelligence
  • AI personas can be used, for example, to progress through a game much faster than an actual player to evaluate game content more quickly; to assess the difficulty of levels with randomness with thousand variations of playthroughs; to generate key performance indicators (KPIs), to increase the speed of design iteration, to free up designers' time to focus on gameplay and high level concepts; to test with the same skill level and style again and again, for example, through various versions an/or iterations of a gaming application 248.
  • KPIs key performance indicators
  • the use of artificial, rather than human intelligence allows the gaming hots 250 to perform with a speed and consistency that cannot practically be performed in the human mind.
  • one or more of the AI personas can operate as regression play -testers that play games based on machine learning on recorded human demonstrations and check that the game is still playable after content or code changes.
  • the regression play- testers can generate a report when errors are found in the game, generate KPIs, predict changes to overall play time and game difficulty and/or operate in conjunction with BEA tools 250 to predict changes to the amount of player behavioral motivation, both positive and negative, including boredom, excitement, completion, etc.
  • the AI personas can work as player stand-ins, AI opponents, and/or NPCs for single and multiplayer games. This allows a game developer to make sure there is always someone to play against and to imitate actual opponents, before and after launch; challenge players with opponents that vary in skill level and style; and generate a living, convincing world with characters that vary in behavioral patterns.
  • the PCG tools 252 use procedural content generation such as procedural content generation via machine learning (PCGML) or other AI to kick-start and accelerate the creative processes of the game developer in the use of the game development application 246 in the development of new gaming applications 248 and/or new content or levels to existing gaming applications.
  • PCGML machine learning
  • the PCG tools 252 can be constructed via a machine learning algorithm and include, for example, Markov random field models, symmetrical Markov random field models, a convolutional neural network, stacking neural networks, a generative adversarial network, deep learning algorithm, un-supervised learning algorithm, Hastings Metropolis sampling or other artificial intelligence model or methodology that is iteratively trained based on the analysis of prior versions of a game, game data 118 such as game telemetry data, behavioral motivation data and/or game play by one or more AI personas and operates to generate new game content such as new game variations, new levels, and other content.
  • game data 118 such as game telemetry data, behavioral motivation data and/or game play by one or more AI personas and operates to generate new game content such as new game variations, new levels, and other content.
  • game playing AI personas can evaluate and critique content generated via PCG by generating AI persona play-traces and statistics across game content and evaluate procedurally generated content in terms of predicted KPIs and/or other performance metrics.
  • This allows the game development application 246 to assist the game developer in understanding and evaluating the play-space of a PCG enabled game, to protect a PCG design from unplayable or degenerate examples.
  • the PCG tools 252 can generate new puzzles, levels or other content by learning from examples provided by the game developer to the game development platform 125 to seed the artificial intelligence model and generate new candidate content for evaluation.
  • the BEA tools 254 operate in conjunction with the game development application 246 to automatically predict player motivations and other player experiences from play traces of players in realtime. Furthermore, the use of BEA tools 254 in combination with gaming hots 250 and/or PCG tools 252 allows a game developer to predict, based on simulated game play, future player motivations and other player experiences from play traces of AI personas. [0040] This use of the game development platform 125 assists the game developer in understanding why players like a particular gaming application 248, reduce churn, optimize player experiences and long-term engagement. In particular, potential game players are different and play for different reasons. Predicting player motivations helps the game developer to understand these differences and groupings across a potential player base.
  • the BEA tools 254 can be constructed via preference learning or other machine learning techniques that are trained based on player questionnaires, game data 118 and/or player data 119 in order to learn and predict actual player motivations. In operation, the BEA tools 254 use game telemetry data, game data 118 and/or player data from other players to predict individual players' reasons for interacting with a game.
  • generating BEA data that indicates to which degree players are motivated by motivation factors allows a game developer to optimize the player experience accordingly, to match players according to their motivations, creating better play sessions, to optimize and individualize games to a player, retaining players and improving life-time value, to identify poor player matches and potential negative interactions before they become a problem, to track developments in your player base over time and know day-by-day if your typical player motivation or behavioral profile starts changing.
  • a game developer is using the game development platform 125 to develop gaming application 248 that is a multiplayer mobile game.
  • the game features two opposing, teams each with up to four characters, playing a form of fantasy American football.
  • the game features 18 characters and in addition to choosing four characters, players can choose between 4 spells they can add to their deck. • This means that the current version of the game supports 293,760 different deck combinations where the properties of the characters and spells could vary indefinitely.
  • the game developer wants to explore the properties of as many different deck solutions and match combinations as possible to optimize the gameplay and ensure a product with high retention that monetizes well, in order to maximize customer lifetime value (LTV).
  • LTV customer lifetime value
  • the gaming hots 250 allow many matches to be executed in parallel, the data aggregated, and compared using statistics, rather requiring qualitative interpretation.
  • Game developers who which to qualitatively inspect a character can do so playing against gaming hots 250, reducing the human labor involved by 50% and freeing employee time for other tasks, while removing the need for scheduling between two employees.
  • gaming hots 250 can be included in the finished game as NPCs to face the player. This removes the need for the game developer to separately develop a player facing AI internally, and improves the game’s hard launch by providing an unlimited number of opponents for new players, as the game developer is building their player base. Case#2
  • the game developer has an existing gaming application 248 that implements a puzzle game.
  • the game developer is developing a gaming application 248 that implements a multiplatform narrative game for PC, Mac and PlayStation 4.
  • the game is a highly complex branching narrative, consisting of around 8 hours worth of gameplay for a full playthrough.
  • the team size is limited and does not have a full time Quality Assurance person on the team.
  • the game developer needs a solution to automatically identify failure points in traversing the story content of the game.
  • Game hots 250 automatically walking through the story of the game, allowing the game developer to identify when the game would crash or the player would get stuck.
  • game hots 250 simulate previous player action to validate that previous demonstrations still are possible following changes to game code or content.
  • Game hots 250 automatically search through be game, walking through the story lines, looking for crash situations and or dead ends.
  • the game development platform 125 can continuously verify that the game works following changes.
  • the game development platform 125 can continuously verify that the game is completable.
  • the game development platform 125 can indefinitely play the game, enabling stress testing that simulates human interactions and provides more realistic use case than simply letting the game run with no input.
  • the game development platform 125 roughly replaces the effort of one QA employee.
  • the game development platform 125 provides improvements in creative efficiency, leading to higher quality content, which can positively impact the final game performance.
  • BEA tools 254 can determine realtime player experience and help improve came completion and player retention.
  • FIG. 3A presents a flow/block diagram representation of a game development pipeline 325 in accordance with an embodiment of the present disclosure.
  • This game development pipeline 325 operates in conjunction with the game development platform 125 of FIGs. 1 and 2 and uses one or more of the functions and features described therewith.
  • a game development pipeline is presented where game development progresses temporally from the initial generation of a game in step 300 through, for example, alpha testing, beta testing and/or soft launch and leading to the generation of an improved game for hard launch in step 314.
  • a game such as an initial version of a gaming application 248 is generated.
  • the initial version of the game is developed by the game developer using the game development application 246, either from scratch or from initial game content generated by PCG tools 252 based on, for example, prior games or prior versions of the game developed by the game developer.
  • step 302 the game is tested using game hots 250 that are non-imitating, e.g. that are developed and trained from testing and evaluation of prior games or prior versions of the game developed by the game developer.
  • the game hots 250 include a library of non-imitating game hots along with descriptive metadata that indicates, for example, the source, prior use, corresponding player motivations and/or other characteristics of each game hot.
  • the game developer can select and evaluate one or more existing game hots that are used for this testing. Once one or more of the game hots 250 is selected, the game can be tested and improved to, for example, identify dead-ends, and begin to balance the game, increase playability, etc.
  • step 304 imitating game hots 250 are generated based on game telemetry data from actual players, such as internal or external players used in testing prior to hard launch.
  • game telemetry data can include data gathered from play traces that can include, for example, game output including audio and pixel data, player input, game status, game events, game achievements, progress toward game goals, game parameters, KPIs and other game analytics.
  • the game hots 250 operate via a machine learning algorithm that is trained via the game telemetry data from actual players.
  • machine learning algorithms include artificial neural networks (or more simply “neural networks” as used herein), support vector machines, Bayesian networks, genetic algorithms and/or other machine learning techniques that are trained via unsupervised, semi-supervised, supervised and/or reinforcement learning and can further include feature learning, sparse dictionary learning, anomaly detection, decision trees, association rules and/or other processes.
  • the game is further tested and improved by monitoring output, such as game telemetry data including, for example, KPIs and other game analytics generated by play of the game by the game hots 250. In this fashion, various versions of the game can be tested, evaluated and improved to, for example, identify dead-ends, further balance the game, further increase playability, optimize predicted player retention, etc.
  • BEA data is gathered from game data, player questionnaires or other experience metrics that includes various player motivations that can be, for example, correlated to KPIs, game events, player behaviors, game status, game achievements, progress toward game goals, game parameters, and other game analytics.
  • Player motivations can be broad motivation factors such as competence, autonomy, relatedness, and presence.
  • player motivations and behaviors can be game-related, including competition, completion, fantasy, destruction, discovery, strategy, excitement, power, including more specific motivations such as achieving a high score, being constantly challenged, being challenged with some other frequency, reaching game goals and achievements, completing levels, relaxing, beating other players or spoiling other players games, cheating, avoiding other players that cheat, and other play styles, etc.
  • the BEA data is used to train one or more BEA tools.
  • the BEA tools 254 can be constructed via preference learning or other ordinal machine learning techniques that are trained based on the BEA data in order to learn and predict actual player motivations.
  • player experiences can be predicted via the BEA tools based on game telemetry data from actual players and/or imitating or non-imitating game hots 350, automatically and in realtime.
  • This player experience data can be used in conjunction with game hot testing in step 306 to further improve the game in step 314 for hard launch, for example, by improving predicted player satisfaction with a game, increasing predicted player retention, and/or increasing predicted revenue generation.
  • the game development pipeline 325 has been described that corresponds to the testing, analysis and refinement of an initial version of the game to an improved game for hard launch, one or more steps in the game development pipeline 325 can also be used to similarly process new versions, updates and/or new content additions to a gaming application 248.
  • the game development pipeline 325 has been described as including step 308 of gathering BEA data and step 310 of generating BEA tools 254 based on the BEA data, in circumstances where the game development platform 125 is used to process similar games, new versions, updates and/or new content additions to a gaming application 248, one or more BEA tools 254 generated from prior versions of the game or from similar games can be selected to for reuse.
  • the BEA tools 254 include a library of BEA tools along with descriptive metadata that indicates, for example, the source, prior use, and/or other characteristics of each BEA tool.
  • the game developer can select and evaluate one or more existing BEA tools 254 that are used in step 312 to predict player experiences including motivations and/or behaviors and other experiences based on game telemetry data from external players.
  • game development platform 125 can fuse innovations on three aspects of a computational model: the input of the model, the computation, and the output of the model.
  • This approach can build on anchoring methods of psychology according to which humans encode values in a comparative (relative) fashion.
  • personas perceive humans (or their demonstrations) via generalizable features and they gradually machine learn to experience the environment as humans would do.
  • the game development platform 125 solves a fundamental problem of psychometrics and human psychology at large: to measure experience computationally in a reliable and valid way. It also addresses a core question of human computer interaction and player experience research: how to simulate experience in simulated worlds the same way humans would feel it. Finally, it solves a traditional problem at the intersection of machine learning and affective computing: how can we learn the most out of less data of a subjective nature.
  • FIG. 3B presents a flow/block diagram 350 representation of a components of the general experience personas in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. 1, 2 and 3 A.
  • This process offers a reliable and effective solution to the generative modelling of player experience (including, for example, motivations and/or behaviors) by combining innovations across the three core subprocesses: the input (descriptor map), the computation per se (generative model), and the output (demonstration).
  • Step 352 - Experience Demonstration the proposed approach for processing the output of the persona is general as it may support any annotation type from traditional psychometrics.
  • experience labels can be collected and processed.
  • human demonstrations of experience can be collected in a continuous fashion via engagement metrics that are extracted from an interaction. That includes the spectrum all the way from the passive observation of a video (e.g. a gameplay video) to the active annotation of any interaction (e.g. a game).
  • Experience labels are processed in an ordinal and unbounded fashion thereby allowing the construction of value-agnostic and general experience models.
  • first-order and second-order combinatorial techniques we can both yield valid and reliable human demonstrations of experience but also generate large datasets from limited data.
  • Questionnaires of any type the dominant state of practice within human computer interaction — are no longer needed (even though questionnaire data can still be processed) and human participation is only limited to realistic small-scale player group sizes.
  • Step 354 Experience Generative Model: experience personas can either learn to predict the experience of a human or even express the experience as a human would do.
  • the game development platform involves methods of deep (preference) learning that learn to predict the global or partial order of labelled experience.
  • the order of human demonstrations defines the utility a reinforcement learning approach (e.g. neuro-evolution, stochastic tree search) will learn to infer. The result is a generative model of experience that is able to “feel” in the simulated environment as a human player would do.
  • Step 356 Experience Descriptor Maps: experience is perceived in the ways interaction is performed and bounded by the experience labelling.
  • the model of perception focuses on areas of labelled experience that are meaningful for the model and eliminates areas that no change is observed or reported with regards to experience.
  • the representation of experience is learned by observing generic aspects of interaction, namely general experience descriptor maps.
  • the design of the maps may vary from high level behavior characterizations to sequential patterns of interaction to detailed latent variables that map to labels of experience. The latter are constructed through simulations of interactions directly when that is possible or indirectly through machine learned forward models of interactions when access to the code that generates the interaction is not available.
  • the BEA tools 254 of the game development platform 125 can be incorporated into the final game itself.
  • individual players can be assessed in terms of their motivations and/or behaviors.
  • a particular game version or game parameter setting can be selected from a library of possible game versions/settings for an individual player in order to complement or otherwise match the particular motivations and/or behaviors predicted to correspond with the individual player in order to, for example, improve the experience for a particular player.
  • a player who likes challenges can be challenged, a player who like completion can be given a game that is easier to complete, etc.
  • the BEA tools 254 of the game development platform 125 can be employed to pair players together in a multiplayer game based on their respective motivations and/or behaviors. For example, a valuable player who, based on a determination by the BEA tools, likes to play the spoiler can be retained by routinely pairing him or her with less- experienced players to foil. In another example, a player, determined to cheat by the BEA tools can be paired with other such players or players who are cheat neutral, avoiding other players who are determined to be demotivated by opposing players who cheat, etc.
  • FIG. 4 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. 1-2, 3 A and 3B.
  • a gaming bot model is generated that corresponds to a gaming bot, such as any of the gaming bots 350 previously described.
  • Step 402 includes receiving game output from a gaming application (app) such as gaming application 348.
  • Step 404 includes generating game input to the gaming app via the gaming bot model, wherein the gaming input corresponds to game play by one or more simulated players.
  • Step 406 includes generating game performance data in response to game play by the simulated player. This game performance data can be used to evaluate game content more quickly; to assess the difficulty of levels with randomness with thousand variations of playthroughs; and can include key performance indicators (KPIs) or other game analytics.
  • KPIs key performance indicators
  • FIG. 5 presents graphical representations 500 and 510 of game telemetry data in accordance with an embodiment of the present disclosure.
  • game telemetry data in the form of actual game output is presented in diagram 500 at a time ti and in diagram 510 at time
  • the game telemetry data includes a character 502 that is generated by a gaming hot model such as a gaming hot 250, another AI persona or other AI.
  • the game telemetry data also includes a character 504 that is generated by an actual player, such as a master player that the gaming hot model is trying to mimic or simulate.
  • the game development application 246 generates the difference between the position of the character 502 and the position of character 504.
  • the difference at time ti, d(ti) is measured as the Euclidean distance between the centroid of characters 502 and 504.
  • the difference at time d(t2) is measured as the Euclidean distance between the centroid of characters 502 and 504.
  • Difference data generated in this fashion can be used as a measure of fit to update the gaming hot to more closely imitate the master player.
  • a gaming hot 250 can use reinforcement learning to learn how to “shadow” the human master player, while also learning from the environment how to cope with new, unseen conditions.
  • a distance measurement from the master to the shadow is used to understand how close it is to replicating the human behavior.
  • values d(ti) can be linear distance measurements, logarithmic distance measurements or distance measurements transformed by some other nonlinear function.
  • Euclidean distances other distances including non-Euclidean distances can likewise be employed.
  • the difference data can include one or more other measurements in addition to or as an alternative to distance, such as the difference in accumulated game score between the gaming hot and the human player during the time period to - tn, the difference in game achievements between the gaming hot and the human player during the time period to - tn, a time difference in reaching a game goal between the gaming hot and the human player during the time period to - tn, a difference in other game metrics or other game analytics between the gaming hot and the human player and/or any combination thereof.
  • FIG. 6 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure. In particular, a method is presented for use with any of the functions and features described in conjunction with FIGs. 1-2, 3 A, 3B, 4 and 5.
  • Step 602 includes generating a gaming hot.
  • Step 604 includes receiving game telemetry data from a gaming app corresponding to an actual player.
  • Step 606 includes generating game telemetry data from the gaming app corresponding to the gaming hot.
  • Step 608 includes updating the gaming hot based on a difference data generated based on the game telemetry data corresponding to an actual player and the game data corresponding to the gaming hot indicating a distance over time between a first character generated by the actual player and a second character generated by the gaming hot.
  • FIG. 7 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. 1-2, 3 A, 3B, and 4-6.
  • Step 702 includes generating behavioral experience analysis (BEA) tools based on preference learning.
  • Step 704 includes receiving game telemetry data from a gaming app.
  • Step 706 includes generating predicted user experiences, such as motivations and/or behaviors, by applying the BEA tools to the game telemetry data.
  • Step 708 includes optimizing the game and/or the player experience based on the predicted user motivations and/or behaviors.
  • BEA behavioral experience analysis
  • FIG. 8 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. FIGs. 1-2, 3 A, 3B, and 4-7.
  • Step 800 includes receiving, via a system including a processor, procedural content based on prior game content.
  • Step 804 includes iteratively improving, via the system, the procedural content based on machine learning and play trace data and/or behavioral motivation data from simulated game play by a gaming hot.
  • Step 806 includes generating, via the system, candidate game content based on the improved procedural content.
  • FIG. 9 presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. FIGs. 1-2, 3 A, 3B, and 4-8.
  • Step 902 includes receiving, via a system including a processor, a gaming application corresponding to a game.
  • Step 904 includes updating the gaming application, via the system, based on a play of the game by at least one non-imitating game hot to generate a first updated gaming application corresponding to a first updated game.
  • Step 906 includes generating, via the system, at least one imitating game bot based on first game telemetry data generated in response to a play of the first updated game by a first plurality of actual players.
  • Step 908 includes generating, via the system, behavioral experience analysis (BEA) data based on the play of the first updated game by the first plurality of actual players.
  • Step 910 includes generating, via the system, at least one BEA tool based on the BEA data.
  • Step 912 includes updating the first gaming application, via the system, based on play of the first updated game by the at least one imitating game bot to generate a second updated gaming application corresponding to a second updated game.
  • BEA behavioral experience analysis
  • Step 914 includes generating predicted player experiences, via the system, based on second telemetry data generated in response to a play of the second updated game by a second plurality of actual players.
  • Step 916 includes updating the second gaming application, via the system, based the predicted player experiences to generate a third updated gaming application corresponding to a third updated game.
  • FIGs. 10A and 10B present graphs in accordance with embodiments of the present disclosure.
  • PCG tools 252 can generate new puzzles, levels or other content by learning from examples provided by the game developer to the game development platform 125 to seed the artificial intelligence model and generate new candidate content for evaluation.
  • PCG tools 252 can employ machine learning models as content generators for games.
  • PCGML uses machine learning algorithms to learn the appropriate invariants of the content it is trained on, so that the content sampled from the learned model retains the “style” of the content, while introducing variety in new levels.
  • the model can learn the maximal lengths of gaps it can generate to still obtain playable levels, but to also introduce new challenges and scenarios (e.g. combinations of gaps) not found in the original levels.
  • This use of PCG tools 252 allows game developers using the game development platform 125 to increase their productivity with pre-generated puzzles, levels and/or other content; to focus on concepts and important details rather than mundane layouts; to start creating from generated examples instead of a blank canvas, and/or generate content in the style and preferences learned from prior game developer based on the seed examples provided by the game developer to the game development platform 125.
  • symmetrical Markov Random Field (SMRF) models can be used to generate new game content in conjunction with the artificial intelligence of the PCG tools 252 of the game development platform 125.
  • the symmetrical MRFs introduced herein modify the standard MRF neighborhood formulation by including symmetric positions in the graph in addition to local neighbors.
  • SMRFs define different neighbors based on the spatial relationships between variables in the graph, and not on the content. This results in an MRF neighborhood that includes non-local neighbors and is able to capture specific symmetrical relationships in the graph. This makes the methods described herein more suited for generative tasks, where the content of each variable may not be well defined initially.
  • MRFs have been used in image and texture processing tasks such as image denoising, image infilling, and signal reconstruction
  • the methods discussed herein in contrast, present improvements to the field of game content generation including image synthesis, game level generation, and other content synthesis. The result improves the technology of game development via automatic game content generation that is quicker and more accurately imitates the content and features of existing games.
  • Pairwise Markov Property any two non-adjacent variables are conditionally independent
  • a Markov Random Field is defined by a neighborhood structure that defines which variables in the graph are dependent on one another. This essentially models the relationships between a variable and its neighbors.
  • MRFs use a neighborhood of spatially local nodes in the graph.
  • X is the current variable/vertex
  • G s are the dependent variables/vertices
  • O’ s are independent of X as shown in the example graph of FIG. 10 A.
  • the SMRF models employed herein can employ a local MRF neighborhood that uses a neighborhood of spatially local nodes in the example graph of FIG. 10A as well as a global neighborhood potentially including distant symmetric positions in the graph as shown in FIG. 10B.
  • X is the current variable/vertex
  • l’s are the dependent variables/vertices
  • 0’s are independent of X. This shows the network structure with horizontally, vertically, and diagonally symmetric nodes.
  • the SMRF uses non-local neighbors in the network structure
  • the network positioning of the neighboring vertices/variables in relation to the current vertex changes depending on the position of the current vertex in the graph i.e., the local neighbors stay the same, but the symmetric neighbors may be very distant or very near the current vertex depending on its distance from the center of the graph.
  • FIG. IOC presents a flowchart representation of a method in accordance with an embodiment of the present disclosure.
  • a method is presented for use with any of the functions and features described in conjunction with FIGs. 1-2, 3 A, 3B, and 4-9.
  • a method is presented that uses Symmetrical Markov Random Field models to generate new game content in conjunction with the PCG tools 252 or other use of the game development platform 125.
  • Step 1002 includes generating, via image analysis, graphs of existing game content.
  • Step 1004 includes generating a symmetrical Markov random field (SMRF) model, based on the graphs.
  • Step 1006 includes automatically generating, via iterative artificial intelligence, new game content based on the SMRF model.
  • each of the graphs of the existing game content represents positions of a plurality of game elements in the existing game content.
  • the SMRF model can include a conditional probability distribution that indicates subsets of the plurality of game elements that are likely to be positioned within a predetermined distance to one another in the graph.
  • Step 1006 can include: selecting a candidate graph based on frequencies of occurrence of the plurality of game elements; generating an improved candidate graph based on the SMRF model; and generating the new game content based on the improved candidate graph.
  • the improved graph can be generated based on a Metropolis-Hastings sampling or other AI-based technique.
  • Step 1006 can include: generating candidate new game content based on the improved candidate graph; testing the candidate new content via game play utilizing a gaming bot - such as one or more of the gaming bots 250; and accepting the candidate new game content as the new game content, based on the game play utilizing the gaming bot.
  • the gaming bot or bots determine that the new game content is playable (e.g. has no dead-ends, infinite loops, is possible to complete) and furthermore has a predicted user experience that is greater than a user experience threshold, it can be accepted as viable new content.
  • the use of artificial, rather than human intelligence allows the gaming bots 250 to perform with a speed and consistency that cannot practically be performed in the human mind, and further improve the technology of game development.
  • the method takes an image of a level, and converts it to graph, such as a tile-based grid representation of the different game elements in a tile grid array.
  • the image is analyzed via an image analysis such as a computer vision model, pattern recognition technique or other artificial intelligence model that assigns a unique symbol to each type of game element present in each grid square of the image.
  • a section of a level of the game Super Mario Brothers is presented in FIG. 11 A is used for training based on the 13 x 13 graph representation (that can also be referred to as an array or tile grid) presented in conjunction with FIG. 11B.
  • the following symbols are used to represent the indicated types of game elements:
  • sentinel tiles “s” are added to denote the boundaries of the level in the grid. While sentinel boundaries are used on the left and bottom edges of the graph, sentinel tiles could likewise be positioned on the upper and right boundaries as well. Furthermore, tiles corresponding to “empty” elements of the game such as clouds, sky, foliage, etc., can be assigned a common symbol indicating they are not active game elements. In the example above, certain games elements are represented by a common symbol that can represent multiple game element types/subtypes.
  • the “G” represents any of a number of enemy types
  • the “M” represents any of a number of types of power-ups
  • the “?” represents any of a number of elements that can be hidden including coins of different values, power-ups of different types, etc.
  • individual symbols could be used to uniquely represent each specific type of game element.
  • a “M6” could represent a sixth type of power-up
  • a “?C3” could represent a third coin value that is hidden
  • a G9 could represent a ninth type of enemy, etc.
  • this process can be repeated for a number of game images of an existing game in order to build a training data set of tile grids/graphs that represent the existing game.
  • machine learning is used to model the relationship between the different tile types and/or subtypes (either generally or specifically) and their corresponding (e.g. relative) positions in the existing game.
  • a SMRF model is trained on tile grids of the training data set, learning the relationships between the different tile types/subtypes and positions versus neighboring grid positions that may be local and may also include symmetrical and possibly non-local neighbors.
  • the SMRF model indicates subsets of the plurality of game elements that are likely to be positioned within a predetermined distance to one another in the graph and results in a conditional probability distribution that describes which tile types are typically located near each other or relative to one another in specific configurations - as presented by the training data set of tile grids from the existing game.
  • FIG. 11C A section of a level of the game Super Mario Brothers is presented in FIG. 11C is used for training based on the 10 x 12 graph representation presented in conjunction with FIG. 11D.
  • the following symbols are used to represent the indicated types of game elements:
  • the method can then employ an iterative search algorithm such as Metropolis- Hastings sampling or other iterative artificial intelligence technique to generate one or more new levels. For example, the method first randomly fills a tile grid with different tile types/subtypes according to their overall frequency in the training data (e.g., 70% empty tile types, 15% ground types, 10% enemies, 5 % powerups, etc.). The method then randomly chooses two positions in the graph, computes, based on the conditional probabilities of the SRMF model, the likelihood of the tile configurations at those positions, and the likelihood of the tiles if the positions were to be swapped. If the swapped likelihood is higher, the method keeps them swapped.
  • an iterative search algorithm such as Metropolis- Hastings sampling or other iterative artificial intelligence technique
  • PCGML operates to inherently learn invariant patterns from a set of examples of an existing game.
  • the type and scale of patterns captured is largely determined by the underlying machine learning approach, the training data, and its representation.
  • the term “patterns” has different connotations in different domains, but the term generally describe regularities within a given object. In the context of frequent pattern mining, this is taken to mean sets of items, sub-sequences or sub-structures that occur multiple times.
  • game design patterns that can take many shapes, including rather abstract patterns related to the overall game design, but also more fine-grained and visible spatial patterns, that define particular relations between tiles that can be learned by the PCGML.
  • CCS Candy Crush Saga
  • candies three or more candies (tiles) can be matched horizontally or vertically with neighboring candies of the same color. When matched, candies disappear from the board. If there are no obstructing items, this causes existing candies to fall down and fill the resulting gaps, and new candies with random colors to be spawned (usually at the top of the board).
  • the game introduces various constraints, obstacles and objectives that together define each level and thus create puzzles of varying difficulty.
  • the following terms can be used:
  • Global pattern A pattern that can only be identified by looking at spatial structure of all elements on the game board.
  • Local pattern A pattern that can be identified in a small area of the game board.
  • CCS levels often exhibit global patterns commonly considered aesthetically pleasing or interesting to the human eye. Examples include placing items on the board in recognizable shapes or in symmetric arrangements (see e.g., areas of the blue squares of FIG. 12B). There are also levels which repeat smaller, local patterns. A common example is that candies with additional beneficial effects are enclosed by obstructing items, making them harder to use (see e.g., areas of the red and green squares of FIG. 12A). Levels can display both local and global patterns.
  • CCS The global aspect of CCS is different from other applications of PCGML, such as Super Mario brothers, where local tile neighborhoods can be used define many of the primary game elements (gaps, pipes, enemy groups, etc.). For this reason, PCGML is employed on CCS to inherently learn different types of patterns, including, for example, both local and global patterns.
  • GANs conditional generative adversarial networks
  • Models can also be trained on each class separately. Both options aim to strengthen the signals around global patterns that are present in the data.
  • a data-driven approach can be employed to enrich data and used to identify class labels automatically using unsupervised learning. If limited labeled data is available, a corresponding approach with a classifier that encodes a learned bias is also possible.
  • a further strategy to improve a model is to augment the algorithm to ensure it focuses on one or more desired patterns. If domain knowledge is available, this can be done explicitly by modifying the structure of the model to detect the desired patterns. For example, to generate MRF models with symmetry, positions can be added that should be mirrored at a given position as input to the algorithm.
  • Another approach is to feed measures describing desired features (e.g. a symmetry score) to the algorithm, so that recognizing the fitness of an individual based on desired features is facilitated. This can be done, for example, by giving additional inputs to a GAN’s discriminator. While these approaches can generate content exhibiting the desired features, they can be reliant on domain knowledge and the ability to characterize the features numerically.
  • desired features e.g. a symmetry score
  • Another approach is to ensure that the input at least allows the algorithm to make connections between items at the scale of the desired global pattern.
  • An example of this data-driven approach is to add a fully-connected layer as the first layer of the discriminator in a GAN.
  • Another approach is to add the position of each input (e.g., as coordinates) to the input of a neural network. Between these two extremes lie approaches with a learned bias. Such a bias can be learned through labeled samples or adversarial training and then given to the model as an additional input.
  • a third approach is filtering solutions, which is most straightforward if done explicitly, but it is conceivable to learn desired patterns and ways to identify them. Filtering can be executed at different times during the training process. Before training would mean creating a representation that only encodes solutions with the desired global patterns. In case of symmetry, for example, only half of the level could be generated and automatically mirrored to construct the final level. Repairing solutions to adhere to patterns (e.g., through mirroring) is also possible during or after training.
  • a further option for filtering after training is applying a search algorithm to the space of generated content, e.g., a latent vector evolution for GANs. . [0095] Continuing with the example of PCGML for CSS, note that CCS contains approximately 80 game elements with different characteristics.
  • CCS levels rely on unique mechanics or game elements, which makes them difficult to replicate. For this reason, a subset of published CCS levels can be selected for training that are more homogeneous. In particular, levels were selected from a specific game mode (Jelly) and levels were discarded containing complex dynamic elements such as frogs and conveyor belts, resulting in the 504 levels used for training. There are still 51 unique items present in the reduced set of levels. Some items can be stacked on the same cell in the board, and 789 unique item stacks were present in this training data set.
  • SHAPE Indicates which game board cell are non-void.
  • REGULAR The six types of regular candy that can be matched with other candies of the same color.
  • each level is represented as a matrix with dimensions 9 x 9 x 6 with a binary encoding to represent the occurrence of an item category in a given cell.
  • the following post-processing method can be introduced. The first four layers cannot coexist in the same cell. The choice for each cell is determined by selecting the layer with highest value. However, if none of the values is higher than a threshold (0.5 for example), the cell is indicated empty. Further, only allow locks to be placed on cells that are not void or empty and jelly is only placed on cells that are not void.
  • a first GAN model discriminator employs filters that produce 9 x 9 patches — in the case of CCS is the entire level.
  • a first variant is trained on the full set of levels (Global GAN), and a second variant is trained on only vertically symmetric levels (GlobalGAN-vert).
  • GlobalGAN generates average scores across the board, however a definite improvement in vertical symmetry is achieved by training only on vertically symmetric levels using (GlobalGAN-vert). However, the score improvement is not major, which may be because most of the original levels are already vertically symmetric.
  • gaming hots can further validate candidate levels generated in this fashion, by determining that the new game content is playable and has a predicted user experience that is greater than a user experience threshold.
  • This use of artificial, rather than human intelligence, to perform various elements of this process allows the PCG tools 252 to perform with a speed, accuracy and consistency that cannot practically be performed in the human mind. While several of the examples have been described above in terms of a single process that is based on conditional tile type probabilities, the process above can be repeated in a hierarchical fashion to first consider tile types and then perform an expanded sampling based on subtypes of selected tile types. Furthermore, AI bots can be used to evaluate particular subtypes of selected tile types in order to enhance or maximize predicted measures of viewer/user experience.
  • the terms “substantially” and “approximately” provides an industry-accepted tolerance for its corresponding term and/or relativity between items.
  • an industry-accepted tolerance is less than one percent and, for other industries, the industry-accepted tolerance is 10 percent or more.
  • Other examples of industry- accepted tolerance range from less than one percent to fifty percent.
  • Industry-accepted tolerances correspond to, but are not limited to, component values, integrated circuit process variations, temperature variations, rise and fall times, thermal noise, dimensions, signaling errors, dropped packets, temperatures, pressures, material compositions, and/or performance metrics.
  • tolerance variances of accepted tolerances may be more or less than a percentage level (e.g., dimension tolerance of less than +/- 1%). Some relativity between items may range from a difference of less than a percentage level to a few percent. Other relativity between items may range from a difference of a few percent to magnitude of differences.
  • the term(s) “configured to”, “operably coupled to”, “coupled to”, and/or “coupling” includes direct coupling between items and/or indirect coupling between items via an intervening item (e.g., an item includes, but is not limited to, a component, an element, a circuit, and/or a module) where, for an example of indirect coupling, the intervening item does not modify the information of a signal but may adjust its current level, voltage level, and/or power level.
  • inferred coupling i.e., where one element is coupled to another element by inference
  • the term “configured to”, “operable to”, “coupled to”, or “operably coupled to” indicates that an item includes one or more of power connections, input(s), output(s), etc., to perform, when activated, one or more its corresponding functions and may further include inferred coupling to one or more other items.
  • the term “associated with”, includes direct and/or indirect coupling of separate items and/or one item being embedded within another item.
  • the term “compares favorably”, indicates that a comparison between two or more items, signals, etc., provides a desired relationship. For example, when the desired relationship is that signal 1 has a greater magnitude than signal 2, a favorable comparison may be achieved when the magnitude of signal 1 is greater than that of signal 2 or when the magnitude of signal 2 is less than that of signal 1.
  • the term “compares unfavorably”, indicates that a comparison between two or more items, signals, etc., fails to provide the desired relationship.
  • one or more claims may include, in a specific form of this generic form, the phrase “at least one of a, b, and c” or of this generic form “at least one of a, b, or c”, with more or less elements than “a”, “b”, and “c”. In either phrasing, the phrases are to be interpreted identically.
  • “at least one of a, b, and c” is equivalent to “at least one of a, b, or c” and shall mean a, b, and/or c. As an example, it means: “a” only, “b” only, “c” only, “a” and “b”, “a” and “c”, “b” and “c”, and/or “a”, “b”, and “c”.
  • processing module may be a single processing device or a plurality of processing devices.
  • a processing device may be a microprocessor, micro controller, digital signal processor, microcomputer, central processing unit, field programmable gate array, programmable logic device, state machine, logic circuitry, analog circuitry, digital circuitry, and/or any device that manipulates signals (analog and/or digital) based on hard coding of the circuitry and/or operational instructions.
  • the processing module, module, processing circuit, processing circuitry, and/or processing unit may be, or further include, memory and/or an integrated memory element, which may be a single memory device, a plurality of memory devices, and/or embedded circuitry of another processing module, module, processing circuit, processing circuitry, and/or processing unit.
  • a memory device may be a read-only memory, random access memory, volatile memory, non-volatile memory, static memory, dynamic memory, flash memory, cache memory, and/or any device that stores digital information.
  • processing module, module, processing circuit, processing circuitry, and/or processing unit includes more than one processing device, the processing devices may be centrally located (e.g., directly coupled together via a wired and/or wireless bus structure) or may be distributedly located (e.g., cloud computing via indirect coupling via a local area network and/or a wide area network).
  • the processing module, module, processing circuit, processing circuitry and/or processing unit implements one or more of its functions via a state machine, analog circuitry, digital circuitry, and/or logic circuitry
  • the memory and/or memory element storing the corresponding operational instructions may be embedded within, or external to, the circuitry comprising the state machine, analog circuitry, digital circuitry, and/or logic circuitry.
  • the memory element may store, and the processing module, module, processing circuit, processing circuitry and/or processing unit executes, hard coded and/or operational instructions corresponding to at least some of the steps and/or functions illustrated in one or more of the Figures.
  • Such a memory device or memory element can be included in an article of manufacture.
  • a flow diagram may include a “start” and/or “continue” indication. The “start” and “continue” indications reflect that the steps presented can optionally be incorporated in or otherwise used in conjunction with one or more other routines.
  • a flow diagram may include an “end” and/or “continue” indication.
  • the “end” and/or “continue” indications reflect that the steps presented can end as described and shown or optionally be incorporated in or otherwise used in conjunction with one or more other routines.
  • start indicates the beginning of the first step presented and may be preceded by other activities not specifically shown.
  • continue indicates that the steps presented may be performed multiple times and/or may be succeeded by other activities not specifically shown.
  • a flow diagram indicates a particular ordering of steps, other orderings are likewise possible provided that the principles of causality are maintained.
  • the one or more embodiments are used herein to illustrate one or more aspects, one or more features, one or more concepts, and/or one or more examples.
  • a physical embodiment of an apparatus, an article of manufacture, a machine, and/or of a process may include one or more of the aspects, features, concepts, examples, etc. described with reference to one or more of the embodiments discussed herein.
  • the embodiments may incorporate the same or similarly named functions, steps, modules, etc. that may use the same or different reference numbers and, as such, the functions, steps, modules, etc. may be the same or similar functions, steps, modules, etc. or different ones.
  • signals to, from, and/or between elements in a figure of any of the figures presented herein may be analog or digital, continuous time or discrete time, and single-ended or differential. For instance, if a signal path is shown as a single-ended path, it also represents a differential signal path. Similarly, if a signal path is shown as a differential path, it also represents a single-ended signal path. While one or more particular architectures are described herein, other architectures can likewise be implemented that use one or more data buses not expressly shown, direct connectivity between elements, and/or indirect coupling between other elements as recognized by one of average skill in the art.
  • module is used in the description of one or more of the embodiments.
  • a module implements one or more functions via a device such as a processor or other processing device or other hardware that may include or operate in association with a memory that stores operational instructions.
  • a module may operate independently and/or in conjunction with software and/or firmware.
  • a module may contain one or more sub- modules, each of which may be one or more modules.
  • a computer readable memory includes one or more memory elements.
  • a memory element may be a separate memory device, multiple memory devices, or a set of memory locations within a memory device.
  • Such a memory device may be a read-only memory, random access memory, volatile memory, non-volatile memory, static memory, dynamic memory, flash memory, cache memory, a quantum register or other quantum memory and/or any other device that stores data in a non-transitory manner.
  • the memory device may be in a form of a solid-state memory, a hard drive memory or other disk storage, cloud memory, thumb drive, server memory, computing device memory, and/or other non-transitory medium for storing data.
  • the storage of data includes temporary storage (i.e., data is lost when power is removed from the memory element) and/or persistent storage (i.e., data is retained when power is removed from the memory element).
  • a transitory medium shall mean one or more of: (a) a wired or wireless medium for the transportation of data as a signal from one computing device to another computing device for temporary storage or persistent storage; (b) a wired or wireless medium for the transportation of data as a signal within a computing device from one element of the computing device to another element of the computing device for temporary storage or persistent storage; (c) a wired or wireless medium for the transportation of data as a signal from one computing device to another computing device for processing the data by the other computing device; and (d) a wired or wireless medium for the transportation of data as a signal within a computing device from one element of the computing device to another element of the computing device for processing the data by the other element of the computing device.
  • a non-transitory computer readable memory is substantially equivalent

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Optimization (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Algebra (AREA)
  • Probability & Statistics with Applications (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Electrically Operated Instructional Devices (AREA)

Abstract

Outil de génération de contenu procédural consistant à : générer, par l'intermédiaire d'une analyse d'image, des graphiques de contenu de jeu existant; générer un modèle de champ aléatoire de Markov symétrique (SMRF) sur la base des graphiques; et générer automatiquement, par l'intermédiaire d'une intelligence artificielle (IA) itérative, un nouveau contenu de jeu sur la base du modèle SMRF.
EP21750106.3A 2020-02-04 2021-01-28 Génération de contenu basée sur l'ia pour des applications de jeu Pending EP4100139A1 (fr)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US202062969928P 2020-02-04 2020-02-04
US17/159,907 US11596867B2 (en) 2019-03-19 2021-01-27 AI-based content generation for gaming applications
PCT/US2021/015496 WO2021158423A1 (fr) 2020-02-04 2021-01-28 Génération de contenu basée sur l'ia pour des applications de jeu

Publications (1)

Publication Number Publication Date
EP4100139A1 true EP4100139A1 (fr) 2022-12-14

Family

ID=77200439

Family Applications (1)

Application Number Title Priority Date Filing Date
EP21750106.3A Pending EP4100139A1 (fr) 2020-02-04 2021-01-28 Génération de contenu basée sur l'ia pour des applications de jeu

Country Status (3)

Country Link
EP (1) EP4100139A1 (fr)
CN (1) CN115175750A (fr)
WO (1) WO2021158423A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116966563A (zh) * 2022-04-29 2023-10-31 完美世界(北京)软件科技发展有限公司 游戏地图中数据交互的方法及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7392390B2 (en) * 2001-12-12 2008-06-24 Valve Corporation Method and system for binding kerberos-style authenticators to single clients
KR101380565B1 (ko) * 2009-12-21 2014-04-02 한국전자통신연구원 실시간 게임 사용자 특성 분석 장치 및 그 제어 방법
JP6103683B2 (ja) * 2012-06-06 2017-03-29 国立大学法人北見工業大学 ロールプレイングゲームの攻略法算出装置、算出方法、算出プログラム及びこのプログラムを記録した記録媒体

Also Published As

Publication number Publication date
CN115175750A (zh) 2022-10-11
WO2021158423A1 (fr) 2021-08-12

Similar Documents

Publication Publication Date Title
JP7399932B2 (ja) クラウドゲーミングネットワークを使用した人工知能（ａｉ）モデルの訓練
Risi et al. Increasing generality in machine learning through procedural content generation
Vinyals et al. Starcraft ii: A new challenge for reinforcement learning
US11325048B2 (en) User experience modeling for gaming applications
Yannakakis Game AI revisited
US10918948B2 (en) Game bot generation for gaming applications
US11826650B2 (en) Multimodal experience modeling for gaming applications
Summerville et al. Learning player tailored content from observation: Platformer level generation from video traces using lstms
US11596867B2 (en) AI-based content generation for gaming applications
US20210366183A1 (en) Glitch detection system
WO2019085823A1 (fr) Procédé, dispositif et support de stockage pour déterminer des informations de jeu, et dispositif électronique
EP4100139A1 (fr) Génération de contenu basée sur l'ia pour des applications de jeu
US11458397B1 (en) Automated real-time engagement in an interactive environment
Wang et al. The fun facets of Mario: Multifaceted experience-driven PCG via reinforcement learning
Ferguson Machine Learning Arena
Togelius et al. Evolutionary Machine Learning and Games
Fuertes et al. Intelligent agents, voice and facial recognition applied in videogames in order to stimulate cognitive development of children—A case study of Tictactoe in 3D
Silva-Galvez et al. A Video Game-Crowdsourcing Approach to Discover a Player’s Strategy for Problem Solution to Housing Development
Correia Modeling and Generation of Playable Scenarios Based on Societies of Agents
Pereira Procedural Generation of Dungeon Maps, Missions and Rooms
CN115944921B (zh) 游戏数据处理方法、装置、设备及介质
Cowley How to advance general game playing artificial intelligence by player modelling
WO2023027922A1 (fr) Assurance de la qualité des joueurs robots pour les applications de jeu
Li et al. Measuring Diversity of Game Scenarios
Campbell et al. Procedural generation of rollercoasters

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20220829

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)