0% Complete
Home
/
13th International Conference on Computer and Knowledge Engineering
Developing Convolutional Neural Networks using a Novel Lamarckian Co-Evolutionary Algorithm
Authors :
Zaniar Sharifi
1
Khabat Soltanian
2
Ali Amiri
3
1- Faculty of Electrical and Computer Engineering University of Zanjan
2- Faculty of Electrical and Computer Engineering University of Zanjan
3- Faculty of Electrical and Computer Engineering University of Zanjan
Keywords :
evolutionary algorithm،evolving deep neural networks،neural architecture search،co-evolution
Abstract :
Neural Architecture Search (NAS) methods autonomously discover high-accuracy neural network architectures, outperforming manually crafted ones. However, The NAS methods require high computational costs due to the high dimension search space and the need to train multiple candidate solutions. This paper introduces LCoDeepNEAT, an instantiation of Lamarckian genetic algorithms, which extends the foundational principles of the CoDeepNEAT framework. LCoDeepNEAT co-evolves CNN architectures and their respective final layer weights. The evaluation process of LCoDeepNEAT entails a single epoch of SGD, followed by the transference of the acquired final layer weights to the genetic representation of the network. In addition, it expedites the process of evolving by imposing restrictions on the architecture search space, specifically targeting architectures comprising just two fully connected layers for classification. Our method yields a notable improvement in the classification accuracy of candidate solutions throughout the evolutionary process, ranging from 2% to 5.6%. This outcome underscores the efficacy and effectiveness of integrating gradient information and evolving the last layer of candidate solutions within LCoDeepNEAT. LCoDeepNEAT is assessed across six standard image classification datasets and benchmarked against eight leading NAS methods. Results demonstrate LCoDeepNEAT’s ability to swiftly discover competitive CNN architectures with fewer parameters, conserving computational resources, and achieving superior classification accuracy compared to other approaches.
Papers List
List of archived papers
Introducing E4MT and LMBNC: Persian pre-processing utilities
Zakieh Shakeri - Mehran Ziabary - Behrooz Vedadian - Fatemeh Azadi - Saeed Torabzadeh - Arian Atefi
T-Rank: Graph Data Analytics for Urban Traffic Modeling
Alireza Safarpour - Iman Gholampour - Amirhossain Aghazadeh Fard - Seyed Mohammad Karbasi
Disturbance Rejection in Quadruple-Tank System by Proposing New Method in Reinforcement Learning
Alireza Nezamzadeh - Mohammadreza Esmaeilidehkordi
SUT: a new multi-purpose synthetic dataset for Farsi document image analysis
Elham Shabaninia - Fatemeh sadat Eslami - Ali Afkari Fahandari - Hossein Nezamabadi-pour
Graph Representation Learning Towards Patents Network Analysis
Mohammad Heydari - Babak Teimourpour
Cross-project Defect Prediction with An Enhanced Transfer Boosting Algorithm
Nazgol Nikravesh - Mohammad Reza Keyvanpour
An Improved and Accurate Measure for Mining Correlated High-utility Itemsets
Amir Masoud Heidari Orojloo - Morteza Keshtkaran
Identification of Botnets and Nodes Attacking Smart Cities by Majority Voting Mechanism and Feature Selection
Maliheh Araghchi - Nazbanoo Farzaneh
The application of Brain Drain Optimization algorithm on static drone placement problem
Mohammad Mehdi Samimi - Alireza Basiri
A Systematic Embedded Software Design Flow for Robotic Applications
Navid Mahdian - Seyed-Hosein Attarzadeh-Niaki - Armin Salimi-Badr
more
Samin Hamayesh - Version 41.5.3