skip to main content
Idioma:

Improving Pre-Trained Weights Through Meta-Heuristics Fine-Tuning

de Rosa, Gustavo H ; Roder, Mateus ; Papa, João Paulo ; Claudio F G dos Santos

arXiv.org, 2022-12

Ithaca: Cornell University Library, arXiv.org

Texto completo disponible

Citas Citado por
  • Título:
    Improving Pre-Trained Weights Through Meta-Heuristics Fine-Tuning
  • Autor: de Rosa, Gustavo H ; Roder, Mateus ; Papa, João Paulo ; Claudio F G dos Santos
  • Materias: Algorithms ; Computer Science - Artificial Intelligence ; Heuristic ; Heuristic methods ; Image classification ; Image reconstruction ; Machine learning ; Multilayer perceptrons ; Multilayers ; Object recognition ; Optimization ; Optimization techniques ; Recurrent neural networks
  • Es parte de: arXiv.org, 2022-12
  • Descripción: Machine Learning algorithms have been extensively researched throughout the last decade, leading to unprecedented advances in a broad range of applications, such as image classification and reconstruction, object recognition, and text categorization. Nonetheless, most Machine Learning algorithms are trained via derivative-based optimizers, such as the Stochastic Gradient Descent, leading to possible local optimum entrapments and inhibiting them from achieving proper performances. A bio-inspired alternative to traditional optimization techniques, denoted as meta-heuristic, has received significant attention due to its simplicity and ability to avoid local optimums imprisonment. In this work, we propose to use meta-heuristic techniques to fine-tune pre-trained weights, exploring additional regions of the search space, and improving their effectiveness. The experimental evaluation comprises two classification tasks (image and text) and is assessed under four literature datasets. Experimental results show nature-inspired algorithms' capacity in exploring the neighborhood of pre-trained weights, achieving superior results than their counterpart pre-trained architectures. Additionally, a thorough analysis of distinct architectures, such as Multi-Layer Perceptron and Recurrent Neural Networks, attempts to visualize and provide more precise insights into the most critical weights to be fine-tuned in the learning process.
  • Editor: Ithaca: Cornell University Library, arXiv.org
  • Idioma: Inglés
Enlaces
Volver a la lista de resultados

  • Realización: ABCD-USP USP   Logos de Redes Sociales: Freepik

Buscando en bases de datos remotas, por favor espere

  • Buscando por
  • enscope:(USP_VIDEOS),scope:("PRIMO"),scope:(USP_FISICO),scope:(USP_EREVISTAS),scope:(USP),scope:(USP_EBOOKS),scope:(USP_PRODUCAO),primo_central_multiple_fe
  • Mostrar lo que tiene hasta ahora