Generation of Sokoban Stages using Recurrent Neural Networks
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2017, Vol 8, Issue 3
Abstract
Puzzles and board games represent several important classes of AI problems, but also represent difficult complexity classes. In this paper, we propose a deep learning based alternative to train a neural network model to find solution states of the popular puzzle game Sokoban. The network trains against a classical solver that uses theorem proving as the oracle of valid and invalid games states, in a setup that is similar to the popular adversarial training framework. Using our approach, we have been able to verify the validity of a Sokoban puzzle up to an accuracy of 99% on the test set. We have also been able to train our network to generate the next possible state of the puzzle board up to an accuracy of 99% on the validation set. We hope that through this approach, a trained neural network will be able to replace human experts and classical rule-based AI in generating new instances and solutions for such games.
Authors and Affiliations
Muhammad Suleman, Farrukh Hasan Syed, Tahir Q. Syed, Saqib Arfeen, Sadaf I. Behlim, Behroz Mirza
Keywords
Related Articles
- EP ID EP251137
- DOI 10.14569/IJACSA.2017.080364
- Views 91
- Downloads 0
An Electronic Intelligent Hotel Management System for International Marketplace
To compete with the international market place, it is crucial for hotel industry to be able to continually improve its services for tourism. In order to construct an electronic marketplace (e-market), it is an inherent r...
OSPF vs EIGRP: A Comparative Analysis of CPU Utilization using OPNET
Routing is difficult in enterprise networks because a packet might have to traverse many intermediary nodes to reach the final destination. The selection of an appropriate routing protocol for a large network is difficul...
BHA-160: Constructional Design of Hash Function based on NP-hard Problem
Secure hash function is used to protect the integrity of the message transferred on the unsecured network. Changes on the bits of the sender’s message are recognized by the message digest produced by the hash function. H...
An Accelerated Architecture Based on GPU and Multi-Processor Design for Fingerprint Recognition
Fingerprint recognition is widely used in security systems to recognize humans. In both industry and scientific literature, many fingerprint identification systems were developed using different techniques and approaches...
Heuristic Evaluation of Serious Game Application for Slow-reading Students
The findings of preliminary studies found that conventional approaches were still relevant but students showed weak and moderate interest and quickly lost focus rather than technology approaches such as serious games wer...