{"title":"利用人工智能加速器训练电脑游戏角色","authors":"Y. Hnatchuk, Yevheniy Sierhieiev, Alina Hnatchuk","doi":"10.31891/csit-2021-3-9","DOIUrl":null,"url":null,"abstract":" A review of the literature has shown that today, given the complexity of computational processes and the high cost of these processes, the gaming computer industry needs to improve hardware and software to increase the efficiency and speed of processing artificial intelligence algorithms. An analysis of existing machine learning tools and existing hardware solutions to accelerate artificial intelligence. A reasonable choice of hardware solutions that are most effective for the implementation of the task. Possibilities of practical use of the artificial intelligence accelerator are investigated. The effectiveness of the proposed solutions has been proven by experiments. The use of an artificial intelligence accelerator model allowed to accelerate the learning of a computer game character by 2.14 times compared to classical methods. \n ","PeriodicalId":353631,"journal":{"name":"Computer systems and information technologies","volume":"51 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"USING ARTIFICIAL INTELLIGENCE ACCELERATORS TO TRAIN COMPUTER GAME CHARACTERS\",\"authors\":\"Y. Hnatchuk, Yevheniy Sierhieiev, Alina Hnatchuk\",\"doi\":\"10.31891/csit-2021-3-9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\" A review of the literature has shown that today, given the complexity of computational processes and the high cost of these processes, the gaming computer industry needs to improve hardware and software to increase the efficiency and speed of processing artificial intelligence algorithms. An analysis of existing machine learning tools and existing hardware solutions to accelerate artificial intelligence. A reasonable choice of hardware solutions that are most effective for the implementation of the task. Possibilities of practical use of the artificial intelligence accelerator are investigated. The effectiveness of the proposed solutions has been proven by experiments. The use of an artificial intelligence accelerator model allowed to accelerate the learning of a computer game character by 2.14 times compared to classical methods. \\n \",\"PeriodicalId\":353631,\"journal\":{\"name\":\"Computer systems and information technologies\",\"volume\":\"51 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computer systems and information technologies\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.31891/csit-2021-3-9\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer systems and information technologies","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.31891/csit-2021-3-9","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
USING ARTIFICIAL INTELLIGENCE ACCELERATORS TO TRAIN COMPUTER GAME CHARACTERS
A review of the literature has shown that today, given the complexity of computational processes and the high cost of these processes, the gaming computer industry needs to improve hardware and software to increase the efficiency and speed of processing artificial intelligence algorithms. An analysis of existing machine learning tools and existing hardware solutions to accelerate artificial intelligence. A reasonable choice of hardware solutions that are most effective for the implementation of the task. Possibilities of practical use of the artificial intelligence accelerator are investigated. The effectiveness of the proposed solutions has been proven by experiments. The use of an artificial intelligence accelerator model allowed to accelerate the learning of a computer game character by 2.14 times compared to classical methods.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
