Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880170
H. Chiang, Ting-Han Wei, I-Chen Wu
This paper improves upon Job-Level (JL) computing, a general distributed computing approach. In JL computing, a client maintains the overall search tree and parcels the overall search into coarse-grained jobs, which are then each calculated by pre-existing game-playing programs. In order to support large-scale problems such as solving 7×7 killall-Go, or building opening books for 9×9 Go or Connect6, JL computing is modified so that the entire search tree is stored in a database, as opposed to simply being stored in the client process' memory. However, the time cost of accessing this database becomes a bottleneck on performance when using a large number of computing resources. This paper proposes a cache mechanism for JL search trees. Instead of the previous approach, where the entire search tree is stored in the database, we maintain parts of the search tree in the memory of the client process to reduce the number of database accesses. Our method significantly improves the performance of job operations. Assuming that each job requires 30 seconds on average, the JL application with this cache mechanism can allow for the use of 5036 distributed computing resources in parallel without database accesses becoming the performance bottleneck.
{"title":"Database caching for job-level computing","authors":"H. Chiang, Ting-Han Wei, I-Chen Wu","doi":"10.1109/TAAI.2016.7880170","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880170","url":null,"abstract":"This paper improves upon Job-Level (JL) computing, a general distributed computing approach. In JL computing, a client maintains the overall search tree and parcels the overall search into coarse-grained jobs, which are then each calculated by pre-existing game-playing programs. In order to support large-scale problems such as solving 7×7 killall-Go, or building opening books for 9×9 Go or Connect6, JL computing is modified so that the entire search tree is stored in a database, as opposed to simply being stored in the client process' memory. However, the time cost of accessing this database becomes a bottleneck on performance when using a large number of computing resources. This paper proposes a cache mechanism for JL search trees. Instead of the previous approach, where the entire search tree is stored in the database, we maintain parts of the search tree in the memory of the client process to reduce the number of database accesses. Our method significantly improves the performance of job operations. Assuming that each job requires 30 seconds on average, the JL application with this cache mechanism can allow for the use of 5036 distributed computing resources in parallel without database accesses becoming the performance bottleneck.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124360748","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7932314
Hongzhi Yin, B. Cui, Hua Lu, Lei Zhao
The peer-review process is the most widely accepted standard for validating products of researchers within the scientific community. It is also adopted by funding agencies. An essential component of peer-review is to find a certain number of experts to review a research paper or a grant proposal. Previous work mainly focuses on finding experts with the necessary expertise relevant to the paper or proposal while ignoring the diversity in the selected reviewers, which potentially leads to the conflict of interest (COI). In this paper, we propose a novel and unified framework that takes three major key factors into account for reviewer assignment: importance, diversity and expertise coverage of a group of reviewers. Our framework selects a panel of reviewers that not only cover all topics of a submission but also reduce various potential COIs. The proposed framework effectively integrates probabilistic topic model and activation spread model in the presence of a social network of researchers. To the best of our knowledge, this is the first work to study the diversity of reviewers and leverage its effect in the reviewer assignment. We conduct extensive experiments to evaluate the performance of our proposed framework for reviewer assignment. The experimental results show that our approach is very effective in finding panels of relevant, authoritative and diverse reviewers for given submissions to review.
{"title":"Expert team finding for review assignment","authors":"Hongzhi Yin, B. Cui, Hua Lu, Lei Zhao","doi":"10.1109/TAAI.2016.7932314","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7932314","url":null,"abstract":"The peer-review process is the most widely accepted standard for validating products of researchers within the scientific community. It is also adopted by funding agencies. An essential component of peer-review is to find a certain number of experts to review a research paper or a grant proposal. Previous work mainly focuses on finding experts with the necessary expertise relevant to the paper or proposal while ignoring the diversity in the selected reviewers, which potentially leads to the conflict of interest (COI). In this paper, we propose a novel and unified framework that takes three major key factors into account for reviewer assignment: importance, diversity and expertise coverage of a group of reviewers. Our framework selects a panel of reviewers that not only cover all topics of a submission but also reduce various potential COIs. The proposed framework effectively integrates probabilistic topic model and activation spread model in the presence of a social network of researchers. To the best of our knowledge, this is the first work to study the diversity of reviewers and leverage its effect in the reviewer assignment. We conduct extensive experiments to evaluate the performance of our proposed framework for reviewer assignment. The experimental results show that our approach is very effective in finding panels of relevant, authoritative and diverse reviewers for given submissions to review.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"225 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122468638","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880111
Jia-Hao Hou, Tsaipei Wang
Ever since its introduction, Monte Carlo Tree Search (MCTS) has shown very good performances on a number of games, most of which are turn-based zero-sum games. More recently, researchers have also started to expand the application of MCTS to other types of games. This paper proposes a new framework of applying MCTS to the game of simulated car racing. We choose to build the search tree in a discretized game-state space and then determine the action from the selected target game state. This allows us to avoid the need to discretize the action space. In addition, we are able to incorporate some heuristics on driving strategies naturally. The resulting controller shows very competitive performance in the open-source racing game TORCS.
{"title":"The development of a simulated car racing controller based on Monte-Carlo tree search","authors":"Jia-Hao Hou, Tsaipei Wang","doi":"10.1109/TAAI.2016.7880111","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880111","url":null,"abstract":"Ever since its introduction, Monte Carlo Tree Search (MCTS) has shown very good performances on a number of games, most of which are turn-based zero-sum games. More recently, researchers have also started to expand the application of MCTS to other types of games. This paper proposes a new framework of applying MCTS to the game of simulated car racing. We choose to build the search tree in a discretized game-state space and then determine the action from the selected target game state. This allows us to avoid the need to discretize the action space. In addition, we are able to incorporate some heuristics on driving strategies naturally. The resulting controller shows very competitive performance in the open-source racing game TORCS.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125741217","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}
Pub Date : 1900-01-01DOI: 10.1109/taai.2016.7880106
R. Susskind
The development of science is clear. From 1950 to 1990 we lived in a world of Contention, with as main question: Will Contention between Paradigms lead to a Paradigm Shift? This development is nicely described by Popper (Logic of Scientific Discovery), Kuhn (The Structure of Scientific Revolutions), Lakatos (The Methodology of Scientific Research Programmes), and Feyerabend (Against Method). In the world of Games, this development is seen in the transition from Minimax to Monte Carlo Tree Search (MCTS). Apparently, the successor of Contention is called Disruption. Currently, we live in a world full of disruptions (1990–2030). In the lecture, I will show the current development by Daniel Dennett (Consciousness Explained, 1990), Richard Susskind (The Future of Law, 1998), Nick Bostron (Superintelligence, 2014), and my own thoughts on Intuition is Programmable (Van den Herik, 2016). The latter is extremely well identified by the power of Deep Learning in the Game of Go (congratulations to Aja Huang). Around 2030 we may expect to see a quantum computer in operation. It will not only produce prime numbers, but also give us the solution of the game of chess (draw or a win for White), and thereafter even for Go (i.e., at a later date). Next to game results, we will observe a continuous development: from decisions made by humans to decisions made by computers. Here, moral constraints are important. Examples will be given.
科学的发展是明确的。从1950年到1990年，我们生活在一个争论的世界里，主要问题是:范式之间的争论会导致范式的转变吗?波普尔(《科学发现的逻辑》)、库恩(《科学革命的结构》)、拉卡托斯(《科学研究计划的方法论》)和费耶阿本德(《反对方法》)很好地描述了这一发展。在游戏世界中，这种发展体现在从极大极小到蒙特卡洛树搜索(MCTS)的过渡中。显然，“争夺”的继任者被称为“破坏”。目前，我们生活在一个充满混乱的世界(1990-2030)。在讲座中，我将展示Daniel Dennett(《意识解释》，1990年)、Richard Susskind(《法律的未来》，1998年)、Nick Bostron(《超级智能》，2014年)和我自己对直觉是可编程的(Van den Herik, 2016年)的看法。后者在围棋游戏中的深度学习能力中得到了很好的体现(祝贺Aja Huang)。2030年左右，我们可能会看到量子计算机投入使用。它不仅会产生质数，而且还会给我们提供国际象棋的解决方案(白棋平局或获胜)，甚至是围棋的解决方案(即，在稍后的日期)。在比赛结果的旁边，我们将看到一个持续的发展:从人类的决策到计算机的决策。在这里，道德约束很重要。将给出例子。
{"title":"Keynote3: Contention and disruption","authors":"R. Susskind","doi":"10.1109/taai.2016.7880106","DOIUrl":"https://doi.org/10.1109/taai.2016.7880106","url":null,"abstract":"The development of science is clear. From 1950 to 1990 we lived in a world of Contention, with as main question: Will Contention between Paradigms lead to a Paradigm Shift? This development is nicely described by Popper (Logic of Scientific Discovery), Kuhn (The Structure of Scientific Revolutions), Lakatos (The Methodology of Scientific Research Programmes), and Feyerabend (Against Method). In the world of Games, this development is seen in the transition from Minimax to Monte Carlo Tree Search (MCTS). Apparently, the successor of Contention is called Disruption. Currently, we live in a world full of disruptions (1990–2030). In the lecture, I will show the current development by Daniel Dennett (Consciousness Explained, 1990), Richard Susskind (The Future of Law, 1998), Nick Bostron (Superintelligence, 2014), and my own thoughts on Intuition is Programmable (Van den Herik, 2016). The latter is extremely well identified by the power of Deep Learning in the Game of Go (congratulations to Aja Huang). Around 2030 we may expect to see a quantum computer in operation. It will not only produce prime numbers, but also give us the solution of the game of chess (draw or a win for White), and thereafter even for Go (i.e., at a later date). Next to game results, we will observe a continuous development: from decisions made by humans to decisions made by computers. Here, moral constraints are important. Examples will be given.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"1997 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128213684","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880171
Chien-Cheng Chen, Meng-Fen Chiang
With the development of GPS and the popularity of smart phones and wearable devices, users can easily log their daily trajectories. Prior works have elaborated on mining trajectory patterns from raw trajectories. Trajectory patterns consist of hot regions and the sequential relationships among them, where hot regions refer the spatial regions with a higher density of data points. Note that some hot regions do not have any meaning for users. Moreover, trajectory patterns do not have explicit time information or semantic information. To enrich trajectory patterns, we propose semantic trajectory patterns which are referred to as the moving patterns with spatial, temporal, and semantic attributes. Given a user trajectory, we aim at mining frequent semantic trajectory patterns. Explicitly, we extract the three attributes from a raw trajectory, and convert it into a semantic mobility sequence. Given such a semantic mobility sequence, we propose two algorithms to discover frequent semantic trajectory patterns. The first algorithm, MB (standing for matching-based algorithm), is a naive method to find frequent semantic trajectory patterns. It generates all possible patterns and extracts the occurrence of the patterns from the semantic mobility sequence. The second algorithm, PS (standing for PrefixSpan-based algorithm), is developed to efficiently mine semantic trajectory patterns. Due to the good efficiency of PrefixSpan, algorithm PS will fully utilize the advantage of PrefixSpan. Since the semantic mobility sequence contains three attributes, we need to further transform it into a raw sequence before using algorithm PrefixSpan. Therefore, we propose the SS algorithm (standing for sequence symbolization algorithm) to achieve this purpose. To evaluate our proposed algorithms, we conducted experiments on the real datasets of Google Location History, and the experimental results show the effectiveness and efficiency of our proposed algorithms.
{"title":"Trajectory pattern mining: Exploring semantic and time information","authors":"Chien-Cheng Chen, Meng-Fen Chiang","doi":"10.1109/TAAI.2016.7880171","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880171","url":null,"abstract":"With the development of GPS and the popularity of smart phones and wearable devices, users can easily log their daily trajectories. Prior works have elaborated on mining trajectory patterns from raw trajectories. Trajectory patterns consist of hot regions and the sequential relationships among them, where hot regions refer the spatial regions with a higher density of data points. Note that some hot regions do not have any meaning for users. Moreover, trajectory patterns do not have explicit time information or semantic information. To enrich trajectory patterns, we propose semantic trajectory patterns which are referred to as the moving patterns with spatial, temporal, and semantic attributes. Given a user trajectory, we aim at mining frequent semantic trajectory patterns. Explicitly, we extract the three attributes from a raw trajectory, and convert it into a semantic mobility sequence. Given such a semantic mobility sequence, we propose two algorithms to discover frequent semantic trajectory patterns. The first algorithm, MB (standing for matching-based algorithm), is a naive method to find frequent semantic trajectory patterns. It generates all possible patterns and extracts the occurrence of the patterns from the semantic mobility sequence. The second algorithm, PS (standing for PrefixSpan-based algorithm), is developed to efficiently mine semantic trajectory patterns. Due to the good efficiency of PrefixSpan, algorithm PS will fully utilize the advantage of PrefixSpan. Since the semantic mobility sequence contains three attributes, we need to further transform it into a raw sequence before using algorithm PrefixSpan. Therefore, we propose the SS algorithm (standing for sequence symbolization algorithm) to achieve this purpose. To evaluate our proposed algorithms, we conducted experiments on the real datasets of Google Location History, and the experimental results show the effectiveness and efficiency of our proposed algorithms.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134275331","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}
Pub Date : 1900-01-01DOI: 10.1109/taai.2016.7880105
In 2006, we launched the Language Grid project to realize a distributed language service infrastructure on the Internet. Using the Language Grid, we worked with a nongovernmental organization since 2011 to support knowledge communications between agricultural experts in Japan and farmers in Vietnam via their children. We observed that a large community emerged to efficiently utilize nonmature machine translation technologies. During this experience, by facing different types of difficulties, we gradually came to understand the nature of intercultural collaboration. Problems are wicked and not easily defined because of their nested and open networked origin. Fortunately, multiagent technologies can be applied to model stakeholders and simulate intercultural collaboration so as to predict the difficulties and to prepare a better support systems. In this talk, we provide a brief history of the research and practice as regards intercultural collaboration, which can be seen as a human-aware research on multi-agent system.
{"title":"Keynote2: Intercultural collaboration as a multi-agent system","authors":"","doi":"10.1109/taai.2016.7880105","DOIUrl":"https://doi.org/10.1109/taai.2016.7880105","url":null,"abstract":"In 2006, we launched the Language Grid project to realize a distributed language service infrastructure on the Internet. Using the Language Grid, we worked with a nongovernmental organization since 2011 to support knowledge communications between agricultural experts in Japan and farmers in Vietnam via their children. We observed that a large community emerged to efficiently utilize nonmature machine translation technologies. During this experience, by facing different types of difficulties, we gradually came to understand the nature of intercultural collaboration. Problems are wicked and not easily defined because of their nested and open networked origin. Fortunately, multiagent technologies can be applied to model stakeholders and simulate intercultural collaboration so as to predict the difficulties and to prepare a better support systems. In this talk, we provide a brief history of the research and practice as regards intercultural collaboration, which can be seen as a human-aware research on multi-agent system.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131261771","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}
With the complexity big data platform architectures, data engineer provides the infrastructure with computation and storage resource for data scientist and data analyst. With those supports, data scientists can focus their domain problem and design the intelligence module (e.g., prepare the data, select/train/tune the machine learning modules and validate the result). However, there is still a gap between system engineer team and data scientists/engineers team. For system engineers, they don't have any knowledge about the application domain and the propose of the analytic program. For data scientists/engineers, they don't know the configuration of the computation system, file system and database. Some application performance issues are related with system configurations. Data scientist and data engineer do not have information and knowledge about the system properties. In this paper, we propose a configuration layer with the current big data platform (i.e., Hadoop) and build a configuration recommendation system to collect data, pre-process data. Based on the processed data, we use semi-automatic feature engineer to provide features for data engineers and build the performance model with three different machine learning algorithms (i.e., random forest, gradient boosting machine and support vector regression). With the same two benchmarks (i.e., wordcount and terasort), our recommended configuration archives remarkable improvement than rule of thumb configuration and better than their improvements.
{"title":"BigExplorer: A configuration recommendation system for big data platform","authors":"Chao-Chun Yeh, Jiazheng Zhou, Sheng-An Chang, Xuan-Yi Lin, Yichiao Sun, Shih-Kun Huang","doi":"10.1109/TAAI.2016.7880179","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880179","url":null,"abstract":"With the complexity big data platform architectures, data engineer provides the infrastructure with computation and storage resource for data scientist and data analyst. With those supports, data scientists can focus their domain problem and design the intelligence module (e.g., prepare the data, select/train/tune the machine learning modules and validate the result). However, there is still a gap between system engineer team and data scientists/engineers team. For system engineers, they don't have any knowledge about the application domain and the propose of the analytic program. For data scientists/engineers, they don't know the configuration of the computation system, file system and database. Some application performance issues are related with system configurations. Data scientist and data engineer do not have information and knowledge about the system properties. In this paper, we propose a configuration layer with the current big data platform (i.e., Hadoop) and build a configuration recommendation system to collect data, pre-process data. Based on the processed data, we use semi-automatic feature engineer to provide features for data engineers and build the performance model with three different machine learning algorithms (i.e., random forest, gradient boosting machine and support vector regression). With the same two benchmarks (i.e., wordcount and terasort), our recommended configuration archives remarkable improvement than rule of thumb configuration and better than their improvements.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"194 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129537815","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880160
Gunarto Sindoro Njoo, C. Lai, Kuo-Wei Hsu
Inferring activities on smartphones is a challenging task. Prior works have elaborated on using sensory data from built-in hardware sensors in smartphones or taking advantage of location information to understand human activities. In this paper, we explore two types of data on smartphones to conduct activity inference: 1) Spatial-Temporal: reflecting daily routines from the combination of spatial and temporal patterns, 2) Application: perceiving specialized apps that assist the user's activities. We employ multi-view learning model to accommodate both types of data and use weighted linear kernel model to aggregate the views. Note that since resources of smartphones are limited, activity inference on smartphones should consider the constraints of resources, such as the storage, energy consumption, and computation power. Finally, we compare our proposed method with several classification methods on a real dataset to evaluate the effectiveness and performance of our method. The experimental results show that our approach outperforms other methods regarding the balance between accuracy, running time, and storage efficiency.
{"title":"Exploring multi-view learning for activity inferences on smartphones","authors":"Gunarto Sindoro Njoo, C. Lai, Kuo-Wei Hsu","doi":"10.1109/TAAI.2016.7880160","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880160","url":null,"abstract":"Inferring activities on smartphones is a challenging task. Prior works have elaborated on using sensory data from built-in hardware sensors in smartphones or taking advantage of location information to understand human activities. In this paper, we explore two types of data on smartphones to conduct activity inference: 1) Spatial-Temporal: reflecting daily routines from the combination of spatial and temporal patterns, 2) Application: perceiving specialized apps that assist the user's activities. We employ multi-view learning model to accommodate both types of data and use weighted linear kernel model to aggregate the views. Note that since resources of smartphones are limited, activity inference on smartphones should consider the constraints of resources, such as the storage, energy consumption, and computation power. Finally, we compare our proposed method with several classification methods on a real dataset to evaluate the effectiveness and performance of our method. The experimental results show that our approach outperforms other methods regarding the balance between accuracy, running time, and storage efficiency.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126955574","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880173
Chia-Chuan Chang, Ting-Han Wei, I-Chen Wu
Recently, Wu et al. introduced a general approach based on distributed computing named Job-Level (JL) Computing. JL Computing has been successfully used to construct the opening books of game-playing programs.? In order to support large-scale computing problems such as solving 7×7 killall-Go, or building opening books for 9×9 Go or Connect6, record databases are used to store JL computing results. In this paper, we further design a mechanism that combines the JL computing system with BOINC (Berkeley Open Infrastructure for Network Computing), so that we can leverage more computing power from volunteers to solve even larger problems. A preliminary experiment has been done to demonstrate the feasibility of the design.
{"title":"Job-level computing with BOINC support","authors":"Chia-Chuan Chang, Ting-Han Wei, I-Chen Wu","doi":"10.1109/TAAI.2016.7880173","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880173","url":null,"abstract":"Recently, Wu et al. introduced a general approach based on distributed computing named Job-Level (JL) Computing. JL Computing has been successfully used to construct the opening books of game-playing programs.? In order to support large-scale computing problems such as solving 7×7 killall-Go, or building opening books for 9×9 Go or Connect6, record databases are used to store JL computing results. In this paper, we further design a mechanism that combines the JL computing system with BOINC (Berkeley Open Infrastructure for Network Computing), so that we can leverage more computing power from volunteers to solve even larger problems. A preliminary experiment has been done to demonstrate the feasibility of the design.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129134111","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}
Pub Date : 1900-01-01DOI: 10.1109/TAAI.2016.7880180
W. Hsu, Ying-ping Chen
In numerous different types of games, the real-time strategy (RTS) ones have always been the focus of gaming competitions, and in this regard, StarCraft can arguably be considered a classic real-time strategy game. Currently, most of the artificial intelligence (AI) players for real-time strategy games cannot reach or get close to the same intelligent level of their human opponents. In order to enhance the ability of Al players and hence improve the playability of games, in this study, we make an attempt to develop for StarCraft a mechanism learning to select an appropriate action to take according to the circumstance. Our empirical results show that action selection can be learned by AI players with the optimization capability of genetic algorithms and that cooperation among identical and/or different types of units is observed. The potential future work and possible research directions are discussed. The developed source code and the obtained results are released as open source.
{"title":"Learning to select actions in starcraft with genetic algorithms","authors":"W. Hsu, Ying-ping Chen","doi":"10.1109/TAAI.2016.7880180","DOIUrl":"https://doi.org/10.1109/TAAI.2016.7880180","url":null,"abstract":"In numerous different types of games, the real-time strategy (RTS) ones have always been the focus of gaming competitions, and in this regard, StarCraft can arguably be considered a classic real-time strategy game. Currently, most of the artificial intelligence (AI) players for real-time strategy games cannot reach or get close to the same intelligent level of their human opponents. In order to enhance the ability of Al players and hence improve the playability of games, in this study, we make an attempt to develop for StarCraft a mechanism learning to select an appropriate action to take according to the circumstance. Our empirical results show that action selection can be learned by AI players with the optimization capability of genetic algorithms and that cooperation among identical and/or different types of units is observed. The potential future work and possible research directions are discussed. The developed source code and the obtained results are released as open source.","PeriodicalId":159858,"journal":{"name":"2016 Conference on Technologies and Applications of Artificial Intelligence (TAAI)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132187055","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：