Proceedings of the Genetic and Evolutionary Computation Conference Companion最新文献_第3页

DynStack

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3533957

A. Beham, Sebastian Leitner, Johannes Karder, Bernhard Werth, S. Wagner

引用次数: 1

A population based-approach to address real-life transport on-demand problems 以人口为基础的方法来解决现实生活中的按需交通问题

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3533940

Sonia Nasri, H. Bouziri, Wassila Aggoune-Mtalaa

引用次数: 0

JGEA

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3533960

Eric Medvet, Giorgia Nadizar, L. Manzoni

引用次数: 8

Policy network for solving flexible job shop scheduling problem with setup times and rescoure constraints 解决具有设置时间和资源约束的柔性作业车间调度问题的策略网络

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3529084

N. Xu, Tian-Ming Bu

{"title":"Policy network for solving flexible job shop scheduling problem with setup times and rescoure constraints","authors":"N. Xu, Tian-Ming Bu","doi":"10.1145/3520304.3529084","DOIUrl":"https://doi.org/10.1145/3520304.3529084","url":null,"abstract":"The flexible job shop scheduling problem (FJSP) is one of the most popular scheduling problems which allows an operation to be completed on several machines in a specific order and is well known as NP-hardness. And in today's industrial manufacturing environment, this problem is becoming more complex. We should take sequence-dependent setup times and resource constraints into account during the manufacturing process. In this article, a policy network is designed to solve the extended FJSP, with the goal of minimizing the total completion time (makespan). Three dispatching rules are proposed to simultaneously select an operation and assign it on a feasible machine at each scheduling point. To describe the manufacturing status at a scheduling stage, six state features are extracted as the input of the network. After the calculation of the network, the distribution of choosing each dispatching rule is obtained. After all of the operations are completed, the policy network is trained using policy gradient, a well-known reinforcement learning (RL) method. Finally, we generate ten benchmark data instances with different scales to compare the performance and efficiency of policy network with other algorithms. The proposed policy network outperforms its competitors, according to the results of the experiments.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128045588","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: 0

The effect of mirrored sampling with active CMA and sample reuse in the CMAES-APOP algorithm 主动CMA镜像采样和样本复用在CMAES-APOP算法中的效果

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3528947

Duc Manh Nguyen

引用次数: 1

Bayesian model selection for reducing bloat and overfitting in genetic programming for symbolic regression 减少符号回归遗传规划中膨胀和过拟合的贝叶斯模型选择

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3528899

G. Bomarito, P. Leser, N. Strauss, K. Garbrecht, J. Hochhalter

{"title":"Bayesian model selection for reducing bloat and overfitting in genetic programming for symbolic regression","authors":"G. Bomarito, P. Leser, N. Strauss, K. Garbrecht, J. Hochhalter","doi":"10.1145/3520304.3528899","DOIUrl":"https://doi.org/10.1145/3520304.3528899","url":null,"abstract":"When performing symbolic regression using genetic programming, overfitting and bloat can negatively impact generalizability and interpretability of the resulting equations as well as increase computation times. A Bayesian fitness metric is introduced and its impact on bloat and overfitting during population evolution is studied and compared to common alternatives in the literature. The proposed approach was found to be more robust to noise and data sparsity in numerical experiments, guiding evolution to a level of complexity appropriate to the dataset. Further evolution of the population resulted not in overfitting or bloat, but rather in slight simplifications in model form. The ability to identify an equation of complexity appropriate to the scale of noise in the training data was also demonstrated. In general, the Bayesian model selection algorithm was shown to be an effective means of regularization which resulted in less bloat and overfitting when any amount of noise was present in the training data. The efficacy of a Genetic Programming (GP) [1] solution is often characterized by its (1) fitness, i.e. ability to perform a training task, (2) complexity, and (3) generalizability, i.e. ability to perform its task in an unseen scenario. Bloat is a common phenomenon for GP in which continued training results in significant increases in complexity with minimal improvements in fitness. There are several theories for the prevalence of bloat in GP which postulate possible evolutionary benefits of bloat [2]; however, for most practical purposes bloat is a hindrance rather than a benefit. For example, bloated solutions are less interpretable and more computationally expensive. Overfitting is another common phenomena in GP and the broader machine learning field. Overfitting occurs when continued training results in better fitness but reduced generalizability.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121453411","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: 7

Evolving convolutional neural networks for intrusion detection system using hybrid multi-strategy aquila optimizer 基于混合多策略aquila优化器的进化卷积神经网络入侵检测系统

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3529021

Wei-Xiang Sun, Qianmu Li, Pengchuan Wang, J. Hou

{"title":"Evolving convolutional neural networks for intrusion detection system using hybrid multi-strategy aquila optimizer","authors":"Wei-Xiang Sun, Qianmu Li, Pengchuan Wang, J. Hou","doi":"10.1145/3520304.3529021","DOIUrl":"https://doi.org/10.1145/3520304.3529021","url":null,"abstract":"As the attack methods of intruders become more complicated and intelligent, the capability of the intrusion detection system (IDS) should be improved to deal with this situation. In recent years, deep learning has been widely used in the IDS because deep learning has advantages in processing high-dimensional data, obtaining hidden information in data, and solving the problem of data imbalance in the network. In addition, swarm intelligence algorithms (SI) as stochastic optimization methods have been extensively employed to improve various optimization problems. In this paper, we propose a hybrid multi-strategy aquila optimizer (HMAO) to obtain the optimal subset of features extracted by convolutional neural networks to train the IDS classifier. We demonstrate the superiority of HMAO by using standard benchmark function experiment. Then, we evaluate the proposed algorithm for IDS on UNSW-NB15 dataset. The experimental results show that HMAO outperforms the original algorithm in terms of the capacity of finding excellent solutions greatly. Furthermore, in the application of IDS, the proposed algorithm performs better than several feature engineering methods from state-of-the-art related works in Accuracy, TPR, FPR, and F-score.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"188 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132329844","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: 0

Evolutionary algorithms meet classical and deep machine learning for skin detection in color images 进化算法满足经典和深度机器学习，用于彩色图像的皮肤检测

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3533941

J. Nalepa, Stanislaw Czembor, Wojciech Dudzik, M. Kawulok

引用次数: 0

Towards the evolutionary assessment of neural transformers trained on source code 基于源代码训练的神经转换器的进化评估

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3534044

Martina Saletta, C. Ferretti

{"title":"Towards the evolutionary assessment of neural transformers trained on source code","authors":"Martina Saletta, C. Ferretti","doi":"10.1145/3520304.3534044","DOIUrl":"https://doi.org/10.1145/3520304.3534044","url":null,"abstract":"In recent years, deep learning have become popular for solving tasks in a wide range of domains. With this growing diffusion, combined with architectures becoming increasingly complex and sophisticated, understanding how deep models make their predictions is now a crucial and challenging research issue. In this work, we are interested in analysing the behaviour of networks that deal with source code, and we address the problem along two experimental directions: in the first one, we study the activations of the neurons of a transformer trained in the detection of software vulnerabilities so as to identify if (and, eventually, where) some human understandable concepts emerge in the network. In the second one, we generate programs by applying a grammar-based evolutionary algorithm with a fitness function that favours individuals which stimulate (or weaken) the activations in neurons where given concepts majorly emerge. We then study the output of the network on sets of such evolved programs, i.e. how the evolutionary pressure along the direction of a concept affects the prediction. We finally discuss how this combination of evolutionary algorithms with fitness functions derived from the neural activations can be effective for explaining the decision process of deep models, and we suggest further research directions.","PeriodicalId":216331,"journal":{"name":"Proceedings of the Genetic and Evolutionary Computation Conference Companion","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133442959","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

NL-SOMA-CLP for real parameter single objective bound constrained optimization 实参数单目标约束优化的NL-SOMA-CLP

Proceedings of the Genetic and Evolutionary Computation Conference Companion Pub Date : 2022-07-09 DOI: 10.1145/3520304.3534051

Hao Ding, Yongfeng Gu, Hua Wu, Jun Zhou

引用次数: 4