Fusing Level and Ruleset Features for Multimodal Learning of Gameplay Outcomes

Liapis, Antonios; Karavolos, Daniel; Makantasis, Konstantinos; Sfikas, Konstantinos; Yannakakis, Georgios N.

doi:10.1109/cig.2019.8848015

Cited by 9 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, experiments in using visualizations of e.g. powerup safety for these FPS level layouts showed that more accurate predictions for gameplay metrics could be achieved [39]. Including this input to the deep learning models of temporal or spatial qualities could also lead to higher accuracy in these harder prediction tasks.…”

Section: Discussionmentioning

confidence: 99%

“…6). This is not surprising, as the predictions for kill ratio (which is used to calculate f ) mainly uses the class parameters [39] with limited impact from the level input. Interestingly, all best differentclass suggestions were with Soldier vs. Sniper (or vice versa); the hypothesis is that both classes are long-range and the range class parameter is a strong predictor of kill ratio.…”

Section: Experiments On Suggestion Generationmentioning

confidence: 99%

“…The level's image is 20×20 pixels, while each character class is described by a vector of 8 parameters (hit points, speed, and 6 properties of the weapon such as rate of fire, accuracy and range). More details on the format of the input can be found in [8], [39].…”

Section: Machine Learning Architecturementioning

confidence: 99%

See 2 more Smart Citations

SuSketch: Surrogate Models of Gameplay as a Design Assistant

Migkotzidis

Liapis

2022

IEEE Trans. Games

Self Cite

View full text Add to dashboard Cite

This paper introduces SuSketch, a design tool for first person shooter levels. SuSketch provides the designer with gameplay predictions for two competing players of specific character classes. The interface allows the designer to work sideby-side with an artificially intelligent creator and to receive varied types of feedback such as path information, predicted balance between players in a complete playthrough, or a predicted heatmap of the locations of player deaths. The system also proactively designs alternatives to the level and class pairing, and presents them to the designer as suggestions that improve the predicted balance of the game. SuSketch offers a new way of integrating machine learning into mixed-initiative co-creation tools, as a surrogate of human play trained on a large corpus of artificial playtraces. A user study with 16 game developers indicated that the tool was easy to use, but also highlighted a need to make SuSketch more accessible and more explainable.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Experiments On Suggestion Generationmentioning

confidence: 99%

See 1 more Smart Citation

SuSketch: Surrogate Models of Gameplay as a Design Assistant

Migkotzidis

Liapis

2022

IEEE Trans. Games

Self Cite

View full text Add to dashboard Cite

show abstract

“…We are aware that bridging the approaches, expectations, and mindsets between technical and nontechnical AI explanations, is a tough task, and needs more research and discussions, not only in the XAI community, but also interdisciplinary collaborations from HCI [51,25,55,15,28], social sciences, and behavioral sciences [21,16,42,30]. Currently, based on our case study and literature research, we are proposing a detailed workflow that put both technical and non-technical AI stakeholders at the center, for high-stake decisions in AI's deployment in real life.…”

Section: Work In Progressmentioning

confidence: 99%

On Two XAI Cultures: A Case Study of Non-technical Explanations in Deployed AI System

Jiang¹,

Senge²

2021

Preprint

View full text Add to dashboard Cite

Explainable AI (XAI) research has been booming, but the question "To whom are we making AI explainable?" is yet to gain sufficient attention. Not much of XAI is comprehensible to non-AI experts, who nonetheless, are the primary audience and major stakeholders of deployed AI systems in practice. The gap is glaring: what is considered "explained" to AI-experts versus non-experts are very different in practical scenarios. Hence, this gap produced two distinct cultures of expectations, goals, and forms of XAI in real-life AI deployments [41]. We advocate that it is critical to develop XAI methods for non-technical audiences. We then present a real-life case study, where AI experts provided non-technical explanations of AI decisions to non-technical stakeholders, and completed a successful deployment in a highly regulated industry. We then synthesize lessons learned from the case, and share a list of suggestions for AI experts to consider when explaining AI decisions to non-technical stakeholders.

show abstract

“…More recently, Li et al (2019) This work draws on convolutional neural networks (CNNs) to predict commentary for a particular frame of a gameplay video. CNNs have been employed to take an input snapshot of a game and predict player experience (Guzdial, Sturtevant, and Li 2016; Liao, Guzdial, and Riedl 2017), game balance (Liapis et al 2019), and the utility of particular game states (Stanescu et al 2016).…”

Section: Related Workmentioning

confidence: 99%

Automated Let's Play Commentary

Shah,

Guzdial,

Riedl

2019

Preprint

View full text Add to dashboard Cite

Let's Plays of video games represent a relatively unexplored area for experimental AI in games. In this short paper, we discuss an approach to generate automated commentary for Let's Play videos, drawing on convolutional deep neural networks. We focus on Let's Plays of the popular game Minecraft. We compare our approach and a prior approach and demonstrate the generation of automated, artificial commentary.

show abstract

Fusing Level and Ruleset Features for Multimodal Learning of Gameplay Outcomes

Cited by 9 publications

References 28 publications

SuSketch: Surrogate Models of Gameplay as a Design Assistant

SuSketch: Surrogate Models of Gameplay as a Design Assistant

On Two XAI Cultures: A Case Study of Non-technical Explanations in Deployed AI System

Automated Let's Play Commentary

Contact Info

Product

Resources

About