Computer Tools in Education最新文献

{"title":"High performance Insights from GPU version CatBoost","authors":"Vasily Ershov","doi":"10.32603/2071-2340-2022-2-59-73","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-2-59-73","url":null,"abstract":"In this paper we discus GPU implementation of open-sourced gradient boosting library CatBoost. This implementation shows the state-of-the-art performance among openly-available libraries and we want to share design insights and used algorithms.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"164 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132517218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modeling for beginners in anydinamics","authors":"Y. Senichenkov","doi":"10.32603/2071-2340-2022-2-74-75","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-2-74-75","url":null,"abstract":"","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"195 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132165822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Routing Algorithm for Vehicles that Avoids Severe Traffic Accident Hotspots on the Road Network (Using the City of Springfield, Massachusetts as a Case Study)","authors":"A. Gershtein, Andrey Terekhov","doi":"10.32603/2071-2340-2021-2-5-18","DOIUrl":"https://doi.org/10.32603/2071-2340-2021-2-5-18","url":null,"abstract":"In [1] statistically significant clusters (hotspots) of severe Traffic Accidents (TA) are found. In this article, as a continuation of [1], a simple routing algorithm to avoid TA hotspots on a road network has been proposed («hotspot avoidance» path). If the road network is represented by a graph with edges and nodes, it is enough to mark every edge which lead to the TA hotspot as «not passable» by letting a attribute of the edge be a very large digit, much greater than max edge length for a given road graph — and the routing algorithm (Dijkstra or Bellman-Ford) will avoid the TA hotspot automatically. Computer simulation was performed for Springfield, MA. It is shown that for the same initial and end points of the route, an average ratio (Route avoiding TA length/Original route length) is bigger for shorter original (without taking into account TA hotspots) routes and gradually slows down to 1.04 for max original route length inside Springfield. Route length ratios show extra route length needed to avoid TA hotspots, but say nothing about new route safety. To estimate safety gain, a new Relative Risk Ratio RRR= (TAs along route which avoids TA hotspots/TAs along original route) was introduced. It is shown for Springfield that relatively short (less than 4 km) original routes are more dangerous (have more TAs along the «hotspot avoidance» route) than original ones, but for relatively long (> 4 km) original routes average RRR gets smaller by 16 % while modified path gets longer by 8 % in average.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131018066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computed multivalues of AGM reveal periodicities of inverse functions","authors":"Franccois Lamarche, H. Ruhland","doi":"10.32603/2071-2340-2022-3-64-81","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-3-64-81","url":null,"abstract":"The article shows how two choices are possible whenever computing the geometric mean, and the repetition of this process can in general yield 2-to-the power N different values when the choices are compounded in the first N steps of evaluation of the arithmeticgeometric mean. This happens not only in the simple AGM involved in the computation of the complete elliptic integral of the first kind, but also in analogous methods for the computation of the complete and incomplete elliptic integrals of the first and second kind.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"AES-11 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132570082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Constructive Problems in Discrete Mathematics: a Comparative Analysis of the Exam with and without a Computer","authors":"A. Chukhnov, S. Pozdnyakov","doi":"10.32603/2071-2340-2022-1-57-84","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-57-84","url":null,"abstract":"The paper analyzes an experiment on the use of manipulators for solving constructive tasks related to the different concepts in teaching discrete mathematics students of a technical university. These concepts include a logical circuit, a finite automaton, a regular expression, a Turing machine, etc. In these problems, it was required to build some construction with given properties. The constructed structures could be tested on the data entered by the user, that is, the possibilities of feedback could be used when searching for a solution. However, the computer tools used --- manipulators --- did not allow checking the correctness of the input solution. The data of the experiment, in which about 100 students participated, are presented. In the experiment, one part of the students was asked to take the exam on a computer with the possibility of using manipulators, while others solved the same tasks «on paper». About half of the tasks were not equipped with manipulators, that is, they were theoretical tasks typical of a written exam. In the proposed set, some constructive and theoretical tasks had a common object, so the study studied the possible impact of constructive activities on the subsequent solution of a theoretical task on the same topic. Also in this experiment, the same problem statement was used for different forms of representation of the mathematical essence, for example, in the terminology of constructing a finite automaton, regular expression, grammar, algorithm. On the basis of statistical processing of the results, conclusions were drawn about a significantly better solving of constructive problems with computer support and about the negative impact of a computer in solving problems that required mental operations that do not have an adequate representation in computer support, for example, on a computer with instrumental support, it is much more effective to solve problems by printed aids. Based on the analysis of the results, some other hypotheses were stated, which are not statistically confirmed, but can serve as a starting point for further research.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130979942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Python 3 Package Developing for Computing in Theory of Multioperation","authors":"Dmitry Eremenko","doi":"10.32603/2071-2340-2022-1-16-29","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-16-29","url":null,"abstract":"This work solves the problem of developing a toolkit that allows computer calculations to obtain new results in the theory of multioperations. The article presents various methods for representing operations and multioperations, and describes algorithms for calculating the superposition of operations and multioperations. Also, the work provides a study of various structures of the Python 3 and the search for the most suitable for the implementation of the representation of operations and multioperations. Based on the results of research on data structures, the architecture of the Python 3 package was developed and implemented for modeling algebras of operations and multioperations in the theory of multioperations.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128810882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Controlling the Movement of an Object on a Field with Barrier Using a Recurrent Neural Network","authors":"A. Lyakhov, D. Korolev","doi":"10.32603/2071-2340-2022-1-5-15","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-5-15","url":null,"abstract":"Consider control model recurrent neural network moving object on a field with barrier using a recurrent neural network. Via genetic algorithm create two neural network different complexity. For each neural network describe algorithm reinforcement learning. Comparison of the effectiveness of their work.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133891621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Overview of Existing Methods for Automatic Generation of Tasks with Conditions in Natural Language","authors":"V. Kruchinin, Vladimir V. Kuzovkin","doi":"10.32603/2071-2340-2022-1-85-96","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-85-96","url":null,"abstract":"The paper considers the main algorithms for generating various school subject problems of closed and open type. Some of these algorythms (i.e. question answering, Visual question answering) use artificial intelligence and some not (i.e. sets of AND/OR tree, templates). It was shown that methods for generating tests using artificial intelligence have a high potential, but they require further development, in particular, the creation of large question-answer database in russian language.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"23 S1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120932352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-agent Architecture for Federated Learning","authors":"Yuleisy Perez Gonzalez, I. Kholod","doi":"10.32603/2071-2340-2022-1-30-45","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-30-45","url":null,"abstract":"The concept of federated learning has become widespread in working with data, mainly due to the fact that it allows training on data directly on the nodes where they are stored. As a result, no data transfer is required. After the training is completed on each node, only the trained model is transmitted to the central server for aggregation. Multi-agent systems behave in a similar way, because agents allow you to train machine learning models on local devices, while preserving confidential information. The ability of agents to interact with each other makes it possible to generalize (aggregate) such models and reuse them. This article presents the architecture of multi-agent systems for federated learning. It highlights the elements that make up the agent platform and the structure of the JADE platform. Describes the lifecycle of all agents used to perform a full training cycle in the MAC_FL environment. The configurations of agent placement for each of the proposed architectures of multi-agent systems of federated learning are analyzed and described: centralized, decentralized and hierarchical.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127342080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigation of the Multiply-add Operation on the Baikal-T Processor","authors":"I. Arkhipov","doi":"10.32603/2071-2340-2022-1-46-56","DOIUrl":"https://doi.org/10.32603/2071-2340-2022-1-46-56","url":null,"abstract":"This article is devoted to the study of the efficiency of the multiply-add operation instruction on the Baikal-T processor. Various examples of using the command are considered, measurements are made and conclusions are formulated in which cases the use of multiply-add operation gives a gain in calculations and in which situations the use of the command is unprofitable in terms of program execution speed.","PeriodicalId":319537,"journal":{"name":"Computer Tools in Education","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116038469","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}