Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
4.5
2.7
Journals
Applied Sciences
Special Issues
Evolutionary Computation Meets Deep Learning
share announcement

Evolutionary Computation Meets Deep Learning

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: 30 July 2024 | Viewed by 6826

Special Issue Editors

Prof. Dr. Yuejiao Gong

E-Mail Website
Guest Editor
School of Computer Science and Technology, South China University of Technology, Guangzhou 510006, China
Interests: evolutionary computation; deep reinforcement learning
Special Issues, Collections and Topics in MDPI journals
Dr. Qiang Yang

E-Mail Website
Guest Editor
School of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing 210044, China
Interests: neural architecture search; swarm intelligence; evolutionary computation
Dr. Ting Huang

E-Mail Website
Guest Editor
Guangzhou Institute of Technology, Xidian University, Guangzhou 510555, China
Interests: evolutionary computation; graph neural network
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Evolutionary computation and deep learning are two mainstream technologies of modern artificial intelligence. They are both biology-inspired computational methods but are engaged in different tasks. Usually, evolutionary algorithms are designed to solve complex optimization problems, whereas deep learning models are built to complete complex learning tasks. Recently, many studies have found that the appropriate combination of these two methods provides rich and flexible ways for the two mature paradigms to boost each other.

The purpose of this Special Issue is to gather a collection of the latest studies on the interplay of evolutionary computation and deep learning, from either theoretical or practical perspectives. We welcome new methods that incorporate different deep learning methods to assist evolutionary algorithms in algorithm configuration, evaluation substitution, etc., as well as the methods that apply different evolutionary algorithms to improve deep learning models in terms of the architectures, training procedures, etc. We invite authors to submit research articles and/or review articles that fit this purpose.

Prof. Dr. Yuejiao Gong
Dr. Qiang Yang
Dr. Ting Huang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • evolutionary computation
  • swarm intelligence
  • deep learning
  • deep reinforcement learning
  • graph neural network
  • neural architecture search

Published Papers (4 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

16 pages, 3934 KiB  
Article
Research on Gearbox Fault Diagnosis Method Based on VMD and Optimized LSTM
by Bang-Cheng Zhang, Shi-Qi Sun, Xiao-Jing Yin, Wei-Dong He and Zhi Gao
Appl. Sci. 2023, 13(21), 11637; https://doi.org/10.3390/app132111637 - 24 Oct 2023
Cited by 2 | Viewed by 808
Abstract
The reliability of gearboxes is extremely important for the normal operation of mechanical equipment. This paper proposes an optimized long short-term memory (LSTM) neural network fault diagnosis method. Additionally, a feature extraction method is employed, utilizing variational mode decomposition (VMD) and permutation entropy [...] Read more.
The reliability of gearboxes is extremely important for the normal operation of mechanical equipment. This paper proposes an optimized long short-term memory (LSTM) neural network fault diagnosis method. Additionally, a feature extraction method is employed, utilizing variational mode decomposition (VMD) and permutation entropy (PE). Firstly, the gear vibration signal is subjected to feature decomposition using VMD. Secondly, PE is calculated as a feature quantity output. Next, it is input into the improved LSTM fault diagnosis model, and the LSTM parameters are iteratively optimized using the chameleon search algorithm (CSA). Finally, the output of the fault diagnosis results is obtained. The experimental results show that the accuracy of the method exceeds 97.8%. Full article
(This article belongs to the Special Issue Evolutionary Computation Meets Deep Learning)
►▼ Show Figures

Figure 1

19 pages, 706 KiB  
Article
A Local Information Perception Enhancement–Based Method for Chinese NER
by Miao Zhang and Ling Lu
Appl. Sci. 2023, 13(17), 9948; https://doi.org/10.3390/app13179948 - 03 Sep 2023
Viewed by 1040
Abstract
Integrating lexical information into Chinese character embedding is a valid method to figure out the Chinese named entity recognition (NER) issue. However, most existing methods focus only on the discovery of named entity boundaries, considering only the words matched by the Chinese characters. [...] Read more.
Integrating lexical information into Chinese character embedding is a valid method to figure out the Chinese named entity recognition (NER) issue. However, most existing methods focus only on the discovery of named entity boundaries, considering only the words matched by the Chinese characters. They ignore the association between Chinese characters and their left and right matching words. They ignore the local semantic information of the character’s neighborhood, which is crucial for Chinese NER. The Chinese language incorporates a significant number of polysemous words, meaning that a single word can possess multiple meanings. Consequently, in the absence of sufficient contextual information, individuals may encounter difficulties in comprehending the intended meaning of a text, leading to the emergence of ambiguity. We consider how to handle the issue of entity ambiguity because of polysemous words in Chinese texts in different contexts more simply and effectively. We propose in this paper the use of graph attention networks to construct relatives among matching words and neighboring characters as well as matching words and adding left- and right-matching words directly using semantic information provided by the local lexicon. Moreover, this paper proposes a short-sequence convolutional neural network (SSCNN). It utilizes the generated shorter subsequence encoded with the sliding window module to enhance the perception of local information about the character. Compared with the widely used Chinese NER models, our approach achieves 1.18%, 0.29%, 0.18%, and 1.1% improvement on the four benchmark datasets Weibo, Resume, OntoNotes, and E-commerce, respectively, and proves the effectiveness of the model. Full article
(This article belongs to the Special Issue Evolutionary Computation Meets Deep Learning)
►▼ Show Figures

Figure 1

15 pages, 3369 KiB  
Article
Deep Reinforcement Learning for Intelligent Penetration Testing Path Design
by Junkai Yi and Xiaoyan Liu
Appl. Sci. 2023, 13(16), 9467; https://doi.org/10.3390/app13169467 - 21 Aug 2023
Cited by 1 | Viewed by 2326
Abstract
Penetration testing is an important method to evaluate the security degree of a network system. The importance of penetration testing attack path planning lies in its ability to simulate attacker behavior, identify vulnerabilities, reduce potential losses, and continuously improve security strategies. By systematically [...] Read more.
Penetration testing is an important method to evaluate the security degree of a network system. The importance of penetration testing attack path planning lies in its ability to simulate attacker behavior, identify vulnerabilities, reduce potential losses, and continuously improve security strategies. By systematically simulating various attack scenarios, it enables proactive risk assessment and the development of robust security measures. To address the problems of inaccurate path prediction and difficult convergence in the training process of attack path planning, an algorithm which combines attack graph tools (i.e., MulVAL, multi-stage vulnerability analysis language) and the double deep Q network is proposed. This algorithm first constructs an attack tree, searches paths in the attack graph, and then builds a transfer matrix based on depth-first search to obtain all reachable paths in the target system. Finally, the optimal path for target system attack path planning is obtained by using the deep double Q network (DDQN) algorithm. The MulVAL double deep Q network(MDDQN) algorithm is tested in different scale penetration testing environments. The experimental results show that, compared with the traditional deep Q network (DQN) algorithm, the MDDQN algorithm is able to reach convergence faster and more stably and improve the efficiency of attack path planning. Full article
(This article belongs to the Special Issue Evolutionary Computation Meets Deep Learning)
►▼ Show Figures

Figure 1

23 pages, 1978 KiB  
Article
Prevention of Controller Area Network (CAN) Attacks on Electric Autonomous Vehicles
by Salah Adly, Ahmed Moro, Sherif Hammad and Shady A. Maged
Appl. Sci. 2023, 13(16), 9374; https://doi.org/10.3390/app13169374 - 18 Aug 2023
Viewed by 1471
Abstract
The importance of vehicle security has increased in recent years in the automotive field, drawing the attention of both the industry and academia. This is due to the rise in cybersecurity threats caused by (1) the increase in vehicle connectivity schemes, such as [...] Read more.
The importance of vehicle security has increased in recent years in the automotive field, drawing the attention of both the industry and academia. This is due to the rise in cybersecurity threats caused by (1) the increase in vehicle connectivity schemes, such as the Internet of Things, vehicle-to-x communication, and over-the-air updates, and (2) the increased impact of such threats because of the added functionalities that are controlled by vehicle software. These causes and threats are further amplified in autonomous vehicles, which are generally equipped with more electronic control units (ECUs) that are connected through controller area networks (CANs). Due to the holistic nature of CANs, attacks on the networks can affect the functionality of all vehicle ECUs and the whole system. This can lead to a breach of privacy, denial of services, alteration of vehicle performance, and exposure to safety threats. Although cryptographic encryption and authentication algorithms and intrusion detection systems (IDS) are currently being used to detect and prevent CAN bus attacks, they have certain limitations. Therefore, this study proposed a mitigation scheme that can detect and prevent such attacks at the ECU level, which could address the limitations of existing algorithms. This study proposed the usage of a secure boot scheme to detect and prevent the execution of malicious codes, as the presence of one or more ECUs with a malicious code is the root cause of most CAN bus attacks. Secure boot schemes apply cryptographic data integrity algorithms to ensure that only authentic and untampered software can run on the vehicle’s ECUs. The selection of an appropriate cryptographic algorithm is important because it affects the secure boot schemes’ security level and performance. Therefore, this study also tested and compared the performance of the proposed secure boot scheme with five different data security algorithms implemented using the hardware security module (HSM) of the TC399 32-bit AURIX™ TriCore™ microcontroller through an electric autonomous vehicle’s control unit. The tests showed that the two most favorable schemes with the selected hardware are the secure boot scheme with the cipher-based message authentication code (CMAC), because it possesses the highest performance with an execution rate of 26.07 (ms/MB), and the secure boot scheme with the elliptic curve digital signature algorithm (ECDSA), because it provides a higher security level with an acceptable compromise in speed. This study also introduced and tested a novel variation of the ECDSA algorithm based on the CMAC algorithm, which was found to have a 19% performance gain over the standard ECDSA-based secure boot scheme. Full article
(This article belongs to the Special Issue Evolutionary Computation Meets Deep Learning)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-4
Back to TopTop