Abstract
To promote effective decision-making in video games and win high scores in a short time, the deep learning algorithms are integrated into game image processing for reinforcement learning. By changing the mapping function from priority to probability, a deep Q-learning priority experience replay algorithm is deduced, which is then compared with the single mapping function. Various researches have proved that the improved algorithm can reproduce the mapping function with higher probability of playback learning of the unit. The advantage of the agent is that it can master the most complete game strategy and ultimately obtain higher scores with the help of the strategy. Therefore, the proposed algorithm is to help the agent formulate a more useful strategy when playing video games. On the one hand, the agent can get better game records. On the other hand, the energy consumed in the game is greatly reduced.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Barhatov V, Campa A, Pletnev D (2018) The impact of internet-technologies development on small business success in Russia. Proc Soc Behav Sci 238:552–561
Barriga Nicolas A, Stanescu Marius, Besoain Felipe et al (2019) Improving RTS game AI by supervised policy learning, tactical search, and deep reinforcement learning. IEEE Comput Intell Mag 14(3):11
Chen XL, Cao L, Li CX et al (2018) Deep reinforcement learning via good choice resampling experience replay memory. Kongzhi Yu Juece control Decis 33(4):600–606
Chiu C, Liu TK, Lu WT et al (2018) A micro-control capture images technology for the finger vein recognition based on adaptive image segmentation. Microsyst Technol 10:1–14
Folkes SR, Lahav O, Maddox SJ (2018) An artificial neural network approach to the classification of galaxy spectra. Mon Not R Astron Soc 283(2):651–665
Geng L, Dong T (2017) An agricultural monitoring system based on wireless sensor and depth learning algorithm. Int J Online Biomed Eng (iJOE) 13(12):127–137
He Y, Zhang Z, Richard Yu F et al (2017) Deep reinforcement learning-based optimization for cache-enabled opportunistic interference alignment wireless networks. IEEE Trans Vehic Technol 66(11):10433–10445
Heo YJ, Kim SJ, Kim D et al (2018) Super-high-purity seed sorter using low-latency image-recognition based on deep learning. IEEE Robot Autom Lett 3(4):3035–3042
Hu Z, Tong H, Zeng Y et al (2018) Fast image recognition of transmission tower based on big data. Prot Control Mod Power Syst 3(1):15
Lcc E, Kutafina E, Jonas SM (2018) Automatic recognition of epileptiform EEG abnormalities. Stud Health Technol Inform 247:171–175
Li Y, Li Y (2017) Face recognition algorithm based on sparse representation of DAE convolution neural network. Recent Patents Comput Sci 10(4):290–298
Li H, Shi Y, Zhang B et al (2018a) superpixel-based feature for aerial image scene recognition. Sensors 18(1):156
Li Y, Zhang HK, Xue XZ et al (2018b) Deep learning for remote sensing image classification: a survey. Wiley Interdiscip Rev Data Min Knowl Discov 12:e1264
Lin M, Sarkar M, Mukherjee A et al (2017) Introspection: accelerating neural network training by learning weight evolution. arXiv preprint arXiv:1704.04959
Liu LB, Hodgins J (2017) Learning to schedule control fragments for physics-based characters using deep Q-learning. Acm Trans Gr 36(4):1
Poon P (2017) Modeling and simulation for exploring power/time trade-off of parallel deep neural network training. Proc Comput Sci 108:2463–2467
Qi X (2018) Rotor resistance and excitation inductance estimation of an induction motor using deep-Q-learning algorithm. Eng Appl Artif Intell 72:67–79
Shen Y, Zhao N, Xia M et al (2017) A deep q-learning network for ship stowage planning problem. Pol Marit Res 24(s3):102–109
Su Y, Liu J, Lin D (2018) Wrong matching points elimination after scale invariant feature transform and its application to image matching. Pattern Recognit Image Anal 28(1):87–96
Wu W, Li AD, He XH et al (2018) A comparison of support vector machines, artificial neural network and classification tree for identifying soil texture classes in southwest China. Comput Electron Agric 144:86–93
Xu H, Han Z, Feng S et al (2018) Foreign object debris material recognition based on convolutional neural networks. Eurasip J Image Video Process 2018(1):21
Zhang Q, Man L, Yang LT et al (2017) Energy-efficient scheduling for real-time systems based on deep Q-learning model. IEEE Trans Sustain Comput 4(99):1
Zhao L, Wang JD, Liu JJ et al (2019) Routing for crowd management in smart cities: a deep reinforcement learning perspective. IEEE Commun Mag 57(4):88–93
Zhou Z, Huang G, Chen H et al (2018) Automatic radar waveform recognition based on deep convolutional denoising auto-encoders. Circuits Syst Signal Process 37(9):4034–4048
Zhu J, Song Y, Jiang D et al (2017) A new deep-Q-learning-based transmission scheduling mechanism for the cognitive internet of things. IEEE Intern Things J 5(4):2375–2385
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
All authors declare that they have no conflict of interest.
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Additional information
Communicated by Mu-Yen Chen.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ji, Z., Xiao, W. Improving decision-making efficiency of image game based on deep Q-learning. Soft Comput 24, 8313–8322 (2020). https://doi.org/10.1007/s00500-020-04820-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-020-04820-z