2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)最新文献

A modified dual-population approach for solving multi-objective problems 求解多目标问题的改进双种群方法

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-12-25 DOI: 10.1109/IESYS.2017.8233567

V. Vu, L. Bui, Trung-Thanh Nguyen

{"title":"A modified dual-population approach for solving multi-objective problems","authors":"V. Vu, L. Bui, Trung-Thanh Nguyen","doi":"10.1109/IESYS.2017.8233567","DOIUrl":"https://doi.org/10.1109/IESYS.2017.8233567","url":null,"abstract":"Maintaining the balance between convergence and diversity plays a vital role in multi-objective evolutionary algorithms (MOEAs). However, most MOEAs cannot reach a satisfying balance, especially when solving problems having complicated pareto optimal sets. In this paper, we present a modified cooperative co-evolution approach for achieving better convergence and diversity simultaneously (namely DPP2). In DPP2, while populations are trying to achieve both criteria, the priority being set for these criteria will be different. One population focuses on achieving better convergence (by using pareto-based ranking scheme), while the other is for ensuring the population diversity (by using the decomposition-based method). After that, we use a cooperation mechanism to integrate the two populations and create a new combined population with hopes of having both characteristics (i.e. converged and diverse). Performance of DPP2 is examined on the well-known benchmarks of multiobjective optimization problems (MOPs) using the hypervolume (HV), the generational distance (GD), the inverted generational distance (IGD) metrics. In comparison with the original version DPP algorithm, experimental results indicated that DPP2 can significantly outperform DPP on the benchmark problems with stable results.","PeriodicalId":429982,"journal":{"name":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","volume":"171 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133634855","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}

引用次数: 1

Convolutional neural networks on assembly code for predicting software defects 基于汇编代码的卷积神经网络预测软件缺陷

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233558

Anh Viet Phan, Minh le Nguyen

引用次数: 28

An evolutionary metaheuristic algorithm to optimise solutions to NES games 一种优化NES游戏解决方案的进化元启发式算法

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233555

Matthew Leane, N. Noman

引用次数: 1

Vietnamese news classification based on BoW with keywords extraction and neural network 基于关键词提取和神经网络的BoW越南语新闻分类

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233559

Toan Pham Van, Ta Minh Thanh

引用次数: 16

Regularizing the loss layer of CNNs for facial expression recognition using crowdsourced labels 基于众包标签的cnn面部表情识别损失层正则化

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233557

Philip Lu, Boyi Li, S. Shama, I. King, Jonathan H. Chan

{"title":"Regularizing the loss layer of CNNs for facial expression recognition using crowdsourced labels","authors":"Philip Lu, Boyi Li, S. Shama, I. King, Jonathan H. Chan","doi":"10.1109/IESYS.2017.8233557","DOIUrl":"https://doi.org/10.1109/IESYS.2017.8233557","url":null,"abstract":"Deep, convolutional neural networks have become the state-of-the-art method for automatic Facial Expression Recognition (FER). Because of the small size and controlled conditions of most FER datasets, however, models can still overfit to the training dataset and struggle to generalize well to new data. We present a novel approach of using crowdsourced label distributions for improving the generalization performance of convolutional neural networks for FER. We implement this as a loss layer regularizer, where the ground truth labels are combined with crowdsourced labels in order to construct a noisy output distribution during training. We use a label disturbance method in which training examples are randomly replaced with incorrect labels drawn from the combined label probability distribution. We compare the performance of our disturbed and undisturbed models in cross-validation testing on the extended Cohn-Kanade dataset and cross-dataset experiments on the MMI, JAFFE, and FER2013 datasets. We find that using our proposed method, test performance is improved on both the MMI and JAFFE datasets. Our results suggest that using nonuniform probability distributions to disturb training can improve generalization performance of CNNs on other FER datasets.","PeriodicalId":429982,"journal":{"name":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126180417","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}

引用次数: 3

An ensemble of rule-based classifiers for incomplete data 不完整数据的基于规则的分类器集合

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233553

Cao Truong Tran, Mengjie Zhang, Peter M. Andreae, Bing Xue, L. Bui

引用次数: 5

Discriminating fish species by an Echo sounder in a set-net using a CNN 利用CNN在设网中利用回声探测仪辨别鱼类

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233571

Yudai Hirama, Soichiro Yokoyama, T. Yamashita, H. Kawamura, Keiji Suzuki, M. Wada

引用次数: 8

Trajectory optimization of a satellite for multiple active space debris removal based on a method for the traveling serviceman problem 基于旅行军人问题方法的多空间碎片主动清除卫星轨迹优化

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233562

Masahiro Kanazaki, Yusuke Yamada, Masashi Nakamiya

{"title":"Trajectory optimization of a satellite for multiple active space debris removal based on a method for the traveling serviceman problem","authors":"Masahiro Kanazaki, Yusuke Yamada, Masashi Nakamiya","doi":"10.1109/IESYS.2017.8233562","DOIUrl":"https://doi.org/10.1109/IESYS.2017.8233562","url":null,"abstract":"Space debris removal is currently a critical issue for space development. It has been reported that five pieces of debris should be removed each year to avoid further increasing the amount of debris in orbit. To remove multiple pieces of debris, one idea is to deliver multiple satellites that can each remove one target debris from orbit. The benefit of this approach is that target debris can be removed without orbit transition, so the satellite can be developed by using simple satellite mechanics. However, multiple satellites need to be launched. Another idea is to use one satellite to remove multiple pieces of space debris. This approach can reduce the launch cost and remove space debris efficiently. However, the satellite must change its orbit after each debris removal, and a technique for optimizing the orbit transition is required. In this study, we focused on the latter strategy and developed a satellite trajectory optimization method for efficient space debris removal. We considered the similarity between the problem of multiple space debris removal and the traveling serviceman problem (TSP) and applied the TSP solution of an evolutionary algorithm (EA) to the former. To improve the efficiency of the multiple debris removal, we maximized the total radar cross-section (RCS), which indicates the amount of space debris, and minimized the total thrust of the satellite. We extended the TSP solution method to multiple objectives by coupling it with a satellite trajectory simulation. To evaluate the developed method, a set of 100 pieces of space debris was selected from a database. The results indicated a tradeoff between the total RCS and total thrust.","PeriodicalId":429982,"journal":{"name":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131606818","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}

引用次数: 9

Evolutionary programming of product design policies. An agent-based model study 产品设计策略的进化规划。基于主体的模型研究

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233552

B. Vermeulen, Bin-Tzong Chie, Shu-Heng Chen, A. Pyka

{"title":"Evolutionary programming of product design policies. An agent-based model study","authors":"B. Vermeulen, Bin-Tzong Chie, Shu-Heng Chen, A. Pyka","doi":"10.1109/IESYS.2017.8233552","DOIUrl":"https://doi.org/10.1109/IESYS.2017.8233552","url":null,"abstract":"The new product development and operational marketing literature grapples with incorporation of uncertainty on the market and technological structure discovered over time. In contrast, market and technological uncertainty is at the heart of neo-Schumpeterian agent-based models used in evolutionary innovation economics. We present a novel agent-based model in which designer agents design products to cater to services desired by user agents. In this model, designers imitate and experiment with design policies with which they engage in contests to puzzle together products. This model thus ‘evolutionary programs’ a commendable design policy for the given market and technological structure. We experimentally vary the segmentation of the market and the density of technological relationships ex ante unknown to designer agents and then study the emerging ‘winning’ design policies. Preliminary simulation results reveal that there is no ‘one-size-fits-all’ design policy, but that winning design policies are tailored to the structure of market and technology following particular rationales. Given that we present a novel model, we critically reflect on the operationalizations and propose further refinements.","PeriodicalId":429982,"journal":{"name":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134607124","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}

引用次数: 2

Agent simulation of functional differentiation Agent模拟功能分化

2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) Pub Date : 2017-11-01 DOI: 10.1109/IESYS.2017.8233564

M. Kubo, Saori Iwanaga, Hiroshi Sato

引用次数: 1