521 Hits in 6.4 sec

An Ameliorated Multiattack Network Anomaly Detection in Distributed Big Data System-based Enhanced Stacking Multiple Binary Classifiers

AbdAllah A. AlHabshy, Bashar I. Hameed, Kamal A. ElDahshan
2022 IEEE Access  
The core binary model is a decision tree classifier with hyperparameters optimized using the grid search method.  ...  Two datasets have been used to validate the experiments, i.e., UNSW-NB15 and CICIDS2017.  ...  These algorithms have been used widely in recent literature. They have been improved, tuned, and optimized for detecting network malicious traffics.  ... 
doi:10.1109/access.2022.3174482 fatcat:xfbsqwyxarhzfnhwaq7hv3rubq

Few-shot Image Classification: Just Use a Library of Pre-trained Feature Extractors and a Simple Classifier [article]

Arkabandhu Chowdhury, Mingchao Jiang, Swarat Chaudhuri, Chris Jermaine
2021 arXiv   pre-print
Recent papers have suggested that transfer learning can outperform sophisticated meta-learning methods for few-shot image classification.  ...  We take this hypothesis to its logical conclusion, and suggest the use of an ensemble of high-quality, pre-trained feature extractors for few-shot image classification.  ...  For our library based methods, this is followed by hyperparameter validation on the CUB birds dataset. After that, each method is tested on the remaining eight datasets without further tuning.  ... 
arXiv:2101.00562v3 fatcat:mhga4gxjmfg3zb5atw67wnakny

Bayesian Hyperparameter Optimization for Deep Neural Network-Based Network Intrusion Detection [article]

Mohammad Masum, Hossain Shahriar, Hisham Haddad, Md Jobair Hossain Faruk, Maria Valero, Md Abdullah Khan, Mohammad A. Rahman, Muhaiminul I. Adnan, Alfredo Cuzzocrea
2022 arXiv   pre-print
Hence, there is a need for an automatic technique to find optimal hyperparameters for the best use of DNN in intrusion detection.  ...  Deep neural networks (DNN) have been successfully applied for intrusion detection problems. The optimal use of DNN-based classifiers requires careful tuning of the hyper-parameters.  ...  Different hyperparameter optimization techniques including random search, meta-heuristic algorithms, and Bayesian optimization are applied to optimize hyperparameters for traditional ML methods like K-Nearest  ... 
arXiv:2207.09902v1 fatcat:zkgssugaeve3vdizo7wv33qvhi

A Deep Learning Based Fault Diagnosis Method with Hyperparameter Optimization by Using Parallel Computing

Chaozhong Guo, Lin Li, Yuanyuan Hu, Jihong Yan
2020 IEEE Access  
This paper proposes an intelligent fault diagnosis method of rolling bearings based on deep belief network (DBN) with hyperparameter optimization by using parallel computing.  ...  INDEX TERMS Deep belief network, hyperparameter optimization, parallel computing, fault diagnosis. 131248 This work is licensed under a Creative Commons Attribution 4.0 License.  ...  In order to obtain the optimal hyperparameters of DBN, GA optimization is used for DBN hyperparameter selection so that the optimal hyperparameters can be found to ensure higher diagnosis accuracy for  ... 
doi:10.1109/access.2020.3009644 fatcat:fo5kvym6jnbxhpt6y34gstqjse

Automated Machine Learning Techniques for Data Streams [article]

Alexandru-Ionut Imbrea
2021 arXiv   pre-print
Moreover, a meta-learning technique for online algorithm selection based on meta-feature extraction is proposed and compared while model replacement and continual AutoML techniques are discussed.  ...  These developments and the continuous-growing demand for machine learning experts led to the development of numerous AutoML tools.  ...  HPO techniques are used to determine the optimal set of hyperparameters for a learning algorithm, while Auto FE aims to extract and select features automatically.  ... 
arXiv:2106.07317v1 fatcat:2phpd2s6j5fpre455jpvrhj5ce

Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions [article]

Hansika Hewamalage, Christoph Bergmeir, Kasun Bandara
2019 arXiv   pre-print
We present an extensive empirical study and an open-source software framework of existing RNN architectures for forecasting, that allow us to develop guidelines and best practices for their use.  ...  For example, we conclude that RNNs are capable of modelling seasonality directly if the series in the dataset possess homogeneous seasonal patterns, otherwise we recommend a deseasonalization step.  ...  Acknowledgment This research was supported by the Australian Research Council under grant DE190100045, Facebook Statistics for Improving Insights and Decisions research award, Monash University Graduate  ... 
arXiv:1909.00590v3 fatcat:ztmnyo6wjrhctmifzu52fia3ba

Detecting Malware Families and Subfamilies using Machine Learning Algorithms: An Empirical Study

Esraa Odat, Batool Alazzam, Qussai M. Yaseen
2022 International Journal of Advanced Computer Science and Applications  
The results show that optimization and ensemble approaches are successful in treating dataset issues, with 95% accuracy in classifying big malware families and 80% in Ransomware subfamilies.  ...  Meta-Multiclass and Random Forest ensemble classifiers are used based on different machine learning classifiers to overcome the imbalance in the data classes.  ...  In addition, we used the genetic algorithm as hyperparameter tuning or parameter tuning.  ... 
doi:10.14569/ijacsa.2022.0130288 fatcat:jtsuubdvx5eejgrnx4sg7p2duu

Condition Based Maintenance in Railway Transportation Systems Based on Big Data Streaming Analysis

Emanuele Fumeo, Luca Oneto, Davide Anguita
2015 Procedia Computer Science  
Streaming Data Analysis (SDA) of Big Data Streams (BDS) for Condition Based Maintenance (CBM) in the context of Rail Transportation Systems (RTS) is a state-of-the-art field of research.  ...  The novelty of our proposal is the heuristic approach for optimizing the trade-off between the accuracy of the OL-SVR models and the computational time and resources needed in order to build them.  ...  Let us take our set of hyperparameters η C , η γ and η which defines the neighborhood set and let us optimize their value based on Algorithm 1.  ... 
doi:10.1016/j.procs.2015.07.321 fatcat:aohuac6p2vetlpr2fhxds5q364

An Efficient Hyperparameter Control Method for a Network Intrusion Detection System Based on Proximal Policy Optimization

Hyojoon Han, Hyukho Kim, Yangwoo Kim
2022 Symmetry  
Through iterative learning using the PPO-based reinforcement learning model, the system is optimized to improve performance automatically according to the network environment, where the IDHCS is used.  ...  learning model based on proximal policy optimization (PPO).  ...  The AE performed hyperparameter optimization using sparsity, unit standardization, orthogonality and a grid search, and the DNN performed intelligent learning rate determination and optimization using  ... 
doi:10.3390/sym14010161 fatcat:pjaujasjwncfrkndwefwhqf3ai

Non-linear Dimensionality Reduction-based Intrusion Detection using Deep Autoencoder

S Sreenivasa Chakravarthi, R. Jagadeesh
2019 International Journal of Advanced Computer Science and Applications  
The performance of the proposed approach is presented and compared with parallel methods used for intrusion detection.  ...  Even though many existing techniques are successfully used for detecting intruders but new variants of malware and attacks are being released every day.  ...  For selecting an optimal set of features various methods including Genetic Algorithm, meta-heuristic algorithms, and Principal Component Analysis (PCA) are used [5] .  ... 
doi:10.14569/ijacsa.2019.0100822 fatcat:4ul6en7fmfbylkytlrqyrb7ciu

Automatic Configuration for Optimal Communication Scheduling in DNN Training [article]

Yiqing Ma, Hao Wang, Yiming Zhang, Kai Chen
2021 arXiv   pre-print
Currently, ByteScheduler adopts Bayesian Optimization (BO) to find the optimal configuration for the hyper-parameters beforehand.  ...  AutoByte extends the ByteScheduler framework with a meta-network, which takes the system's runtime statistics as its input and outputs predictions for speedups under specific configurations.  ...  Parameter configuration is necessary for many applications, such as big data analytics [2] , tuning machine learning hyper-parameters [42] , and databases [15] .  ... 
arXiv:2112.13509v1 fatcat:hgap4lualba3houm7h4j4vrfsq

A predictive model for network intrusion detection using stacking approach

Smitha Rajagopal, Poornima Panduranga Kundapur, Hareesh Katiganere Siddaramappa
2020 International Journal of Electrical and Computer Engineering (IJECE)  
This work presents an ensemble approach for network intrusion detection using a concept called Stacking.  ...  Machine learning algorithms are extensively used to detect intrusions by dint of their remarkable predictive power.  ...  Another algorithm called particle swarm optimization was used to enhance the performance of the classifier by fine tuning the parameters.  ... 
doi:10.11591/ijece.v10i3.pp2734-2741 fatcat:42kjlmklsbfonpgthvlfji5qaq

Machine Learning Methods for Detecting Fraud in Online Marketplaces

Raoul Dekou, Sabljic Savo, Simon Kufeld, Diana Francesca, Ricardo Kawase
2021 International Conference on Information and Knowledge Management  
We found that the stacked ensemble model provides the best performance (F1=0.73) when compared to other commonly used models in the field.  ...  In this context, we leveraged the power of existing open sources machine learning libraries H2O and Catboost and designed a pipeline to collect, process and predict the likelihood of a private seller's  ...  Acknowledgements We would like to thank the Customer Service team at for their countless hours of manual work in detecting fraud, and for providing us the ground truth to start our work.  ... 
dblp:conf/cikm/DekouSKFK21 fatcat:kx3ms2g6unddtimtrxe5ed7pce

Meta-Generalization for Multiparty Privacy Learning to Identify Anomaly Multimedia Traffic in Graynet [article]

Satoshi Kamo, Yiqiang Sheng
2022 arXiv   pre-print
This letter explores meta-generalization for a multiparty privacy learning model in graynet to improve the performance of anomaly multimedia traffic identification.  ...  Experimental results demonstrate that the proposal outperforms the state-of-the-art learning models for identifying anomaly multimedia traffic.  ...  CONCLUSION A novel algorithm for multiparty privacy learning in graynet was proposed by exploring and testing the principles of meta-generalization.  ... 
arXiv:2201.03027v1 fatcat:vnctz7kbhfaunj4nsmib6ms3fq

The Effect of Diversity in Meta-Learning [article]

Ramnath Kumar, Tristan Deleu, Yoshua Bengio
2022 arXiv   pre-print
For this experiment, we train on multiple datasets, and with three broad classes of meta-learning models - Metric-based (i.e., Protonet, Matching Networks), Optimization-based (i.e., MAML, Reptile, and  ...  In this work, we find evidence to the contrary; we study different task distributions on a myriad of models and datasets to evaluate the effect of task diversity on meta-learning algorithms.  ...  Acknowledgements We would like to thank Sony Corporation for funding this research through the Sony Research Award Program.  ... 
arXiv:2201.11775v2 fatcat:2tg35hucqbdaxjctpbwdrlob7y
« Previous Showing results 1 — 15 out of 521 results