Automated Game Testing with ICARUS

Pfau, Johannes; Smeddinck, Jan David; Malaka, Rainer

doi:10.1145/3130859.3131439

Cited by 42 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Schatten et al [14] simulated large-scale dynamic agent systems to test quest solvability in MMORPGs. Pfau et al [13] introduced a generic adventure solver traversing point-and-click adventure games via reinforcement learning and reporting crashes, dead-ends and performance issues. Van Kreveld et al [29] and Southey et al [30] assessed difficulty or interestingness approximations of levels or mechanics by machine learning of descriptive ingame metrics.…”

Section: Related Workmentioning

confidence: 99%

“…For instance, AI can predict users' high-level behavior [8] (if they will leave the game, if they will make purchases) or their motivation levels [9] based on past trends in the userbase. Specifically with regards to automated playtesting, AI agents have been used to discover bugs or game crashes [10], constraint violations [11], [12] to identify dead-end game states [13] or unreachable states [14]. Notably, most of these agents follow ad-hoc heuristics and constraints and do not necessarily match how players act in their games.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dungeons & Replicants II: Automated Game Balancing Across Multiple Difficulty Dimensions via Deep Player Behavior Modeling

et al. 2023

View full text Add to dashboard Cite

Video game testing has become a major investment of time, labor and expense in the game industry. Particularly the balancing of in-game units, characters and classes can cause long-lasting issues that persist years after a game's launch. While approaches incorporating artificial intelligence have already shown successes in reducing manual effort and enhancing game development processes, most of these draw on heuristic, generalized or optimal behavior routines, while actual low-level decisions from individual players and their resulting playing styles are rarely considered. In this paper, we apply Deep Player Behavior Modeling to turn atomic actions of 213 players from 6 months of single-player instances within the MMORPG Aion into generative models that capture and reproduce particular playing strategies. In a subsequent simulation, the resulting generative agents ("replicants") were tested against common NPC opponent types of MMORPGs that iteratively increased in difficulty, respective to the primary factor that constitutes this enemy type (Melee, Ranged, Rogue, Buffer, Debuffer, Healer, Tank or Group). As a result, imbalances between classes as well as strengths and weaknesses regarding particular combat challenges could be identified and regulated automatically.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dungeons & Replicants II: Automated Game Balancing Across Multiple Difficulty Dimensions via Deep Player Behavior Modeling

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Furthermore, as in the sixth section, we introduced the QA of behaviors of the social robot, from the viewpoint of their functions and the quality of the animation. These perspectives, especially in artistic areas, are explored in computer game industries [28,29] but are still in their infancy in social robotics.…”

Section: Production Of Character Animation In Sociable Robotmentioning

confidence: 99%

Production of Character Animation in a Home Robot: A Case Study of LOVOT

Yoshida

Yonemura

Emoto³

et al. 2021

Int J of Soc Robotics

View full text Add to dashboard Cite

This is a case study focused on the development of LOVOT, a consumer home robot that went on sale in 2019. Since its unveiling in 2018, approximately 20,000 people have visited the LOVOT demonstration event as of the end of July 2019. In practical terms, LOVOT lives at home with the user long-term, similar to a companion animal; LOVOT performs continuous natural motion for user comfort. We applied movements using professional animator techniques based on the principles of traditional animation to LOVOT. We have identified specific practical techniques in home robot animation through iterative prototyping in our synergetic development of character animation techniques and tools. In this paper, we introduce our practical methods to develop many behaviors in a single robotic agent and is the collaborative efforts of a group of people with diverse professional backgrounds. Furthermore, we summarize how traditional techniques of character animation were applied, and new techniques were required from the perspective of professional animators.

show abstract

“…In APT, pre-trained artificial intelligence (AI) agents will "play" the game, test for bugs and provide feedback to game designers regarding balance in game mechanics, and provide QA (Pfau, Smeddinck, & Malaka, 2017).…”

Section: Introductionmentioning

confidence: 99%

Automated playtesting of platformer games using reinforcement learning

Sriram¹

View full text Add to dashboard Cite

Platformer games are popular in the video game industry and their design require much efforts from game companies. As part of the design process, playtesting is key for improving the gameplay before game release. Playtesting is the quality assurance phase of the game development cycle where people are hired to play the game, report bugs and provide feedback regarding the playability of the game. This feedback could be used for game balancing (process of tuning game rules to prevent them from being ineffective or provide undesirable results). However, playtesting may be expensive if done manually and may require several iterations, resulting in high budget requirement and time for game companies. In this thesis, we investigate a way to automatically playtest 2D platformer levels using a combination of deep reinforcement learning and curriculum learning, for both quality assurance and game balancing. Deep Reinforcement Learning has contributed greatly in playing games (Atari and Dota 2) and in this thesis, we will try to replicate the results to playtest games. Curriculum learning is an approach that has shown promising results thus we will explore it to derive useful results. We develop our APT tool by training an artificial intelligence (AI) agent on several different platformer levels following a curriculum, and use the trained agent to playtest newly-created levels. Our APT is able to identify areas of the level that needed design improvements and further gameplay balancing. We contribute a reliable APT tool for designers that wish to easily design 2D platformer games and a discussion of how our results extend to APT at-large.

show abstract

Automated Game Testing with ICARUS

Cited by 42 publications

References 2 publications

Dungeons & Replicants II: Automated Game Balancing Across Multiple Difficulty Dimensions via Deep Player Behavior Modeling

Dungeons & Replicants II: Automated Game Balancing Across Multiple Difficulty Dimensions via Deep Player Behavior Modeling

Production of Character Animation in a Home Robot: A Case Study of LOVOT

Automated playtesting of platformer games using reinforcement learning

Contact Info

Product

Resources

About