Deep learning for procedural content generation

This paper provides a comprehensive overview of the applications of game theory in deep learning. Today, deep learning is a fast-evolving area for research in the domain of artificial intelligence. Alternatively, game theory has been showing its multi-dimensional applications in the last few decades. The application of game theory to deep learning includes another dimension in research. Game theory helps to model or solve various deep learning-based problems. Existing research contributions demonstrate that game theory is a potential approach to improve results in deep learning models. The design of deep learning models often involves a game-theoretic approach. Most of the classification problems which popularly employ a deep learning approach can be seen as a Stackelberg game. Generative Adversarial Network (GAN) is a deep learning architecture that has gained popularity in solving complex computer vision problems. GANs have their roots in game theory. The training of the generators and discriminators in GANs is essentially a two-player zero-sum game that allows the model to learn complex functions. This paper will give researchers an extensive account of significant contributions which have taken place in deep learning using game-theoretic concepts thus, giving a clear insight, challenges, and future directions. The current study also details various real-time applications of existing literature, valuable datasets in the field, and the popularity of this research area in recent years of publications and citations.

show abstract

“…In the future, the content generation framework requires more than one domain of computational creativity within a game-theoretic context [77].…”

Section: Future Research Challengesmentioning

confidence: 99%

Applications of game theory in deep learning: a survey

Hazra

Anjaria

2022

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Even though the models obtained outstanding results on the used benchmarks, the authors explained that experimenting with richer graph structural bias has a significant impact. Liu et al [16] discussed the potential benefits and limitations of procedural content generation in video games depending on deep learning methods. However, further investigations are required for events, goals, and character generations.…”

Section: A Text Generationmentioning

confidence: 99%

A Predictive Text System for Medical Recommendations in Telemedicine: A Deep Learning Approach in the Arabic Context

Habib¹,

Faris²,

Qaddoura

et al. 2021

IEEE Access

View full text Add to dashboard Cite

We are currently witnessing an immense proliferation of natural language processing (NLP) applications. Natural language generation (NLG) has emerged from NLP and is now commonly utilized in various applications, including chatting applications. The objective of this paper is to propose a deep learningbased language generation model that simplifies the process of writing medical recommendations for doctors in an Arabic context, to improve service satisfaction and patient-doctor interactions. The developed language generation model is a predictive text system intended for next word prediction in a telemedicine service. Altibbi a -a digital platform for telemedicine and teleconsultations services in the Middle East and the North Africa (MENA) region-was utilized as a case study for the textual prediction process. The proposed model was trained using data obtained from Altibbi databases related to medical recommendations, particularly gynecology, dermatology, psychiatric diseases, urology, and internist diseases. Variants of deep learning models were implemented and optimized for next word prediction, based on the unidirectional and bidirectional long short-term memory (LSTM and BiLSTM), the one-dimensional convolutional neural network (CONV1D), and a combination of LSTM and CONV1D (LSTM-CONV1D). The algorithms were trained using two versions of the datasets (i.e., 3-gram and 4gram representations) and evaluated in terms of their training accuracy and loss, validation accuracy and loss, and testing accuracy per their matching scores. The proposed models' performances were comparable. CONV1D produced the most promising matching score.

show abstract

“…Procedural Content Generation via Machine Learning (PCGML) [3] was reviewed where various learning methods in addition to different data sources and representations were discussed. In a more recent work, Deep learning for procedural content generation [12] was reviewed and the review also discussed deep learning methods that are potentially useful for PCG but were still not widely used. Another recent work [13] presented a review of PCG with a focus on puzzle generation.…”

Section: Related Workmentioning

confidence: 99%

Procedural Level Generation for Sokoban via Deep Learning: An Experimental Study

Zakaria¹,

Hadhoud²,

Fayek³

2021

Preprint

View full text Add to dashboard Cite

Deep learning for procedural level generation has been explored in many recent works, however, experimental comparisons with previous works are either nonexistent or limited to the works they extend upon. This paper’s goal is to conduct an experimental study on four recent deep learning procedural level generation methods for Sokoban (size = 7 × 7) to explore their strengths and weaknesses and provide insights for possible research directions. The methods will be bootstrapping conditional generative models, controllable & uncontrollable procedural content generation via reinforcement learning (PCGRL) and generative playing networks. We will propose some modifications to either adapt the methods to the task or improve their efficiency and performance. For the bootstrapping method, we propose using diversity sampling to improve the solution diversity, auxiliary targets to enhance the models’ quality and Gaussian mixture models to improve the sample quality. The results show that diversity sampling at least doubles the unique plan count in the generated levels. On average, auxiliary targets increases the quality by 24% and sampling conditions from Gaussian mixture models increases the sample quality by 13%. Overall, PCGRL shows superior quality and diversity while generative adversarial networks exhibit the least control confusion when trained with diversity sampling and auxiliary targets.

show abstract

Deep learning for procedural content generation

Cited by 96 publications

References 99 publications

Applications of game theory in deep learning: a survey

Applications of game theory in deep learning: a survey

A Predictive Text System for Medical Recommendations in Telemedicine: A Deep Learning Approach in the Arabic Context

Procedural Level Generation for Sokoban via Deep Learning: An Experimental Study

Contact Info

Product

Resources

About