Enhancing CNN Classification with Lamarckian Memetic Algorithms and Local Search
It addresses optimization challenges in CNN training for image classification, offering a robust alternative to traditional methods, though it appears incremental as it builds on existing metaheuristic approaches.
This paper tackled the problem of local minima entrapment in gradient-based optimization for deep neural networks by proposing a two-stage training technique with population-based metaheuristic algorithms and local search, resulting in improved accuracy and computational efficiency over state-of-the-art methods like ADAM, especially for complex networks with many parameters.
Optimization is critical for optimal performance in deep neural networks (DNNs). Traditional gradient-based methods often face challenges like local minima entrapment. This paper explores population-based metaheuristic optimization algorithms for image classification networks. We propose a novel approach integrating a two-stage training technique with population-based optimization algorithms incorporating local search capabilities. Our experiments demonstrate that the proposed method outperforms state-of-the-art gradient-based techniques, such as ADAM, in accuracy and computational efficiency, particularly with high computational complexity and numerous trainable parameters. The results suggest that our approach offers a robust alternative to traditional methods for weight optimization in convolutional neural networks (CNNs). Future work will explore integrating adaptive mechanisms for parameter tuning and applying the proposed method to other types of neural networks and real-time applications.