Fractal Few-Shot Learning

Fobao Zhou

doi:10.1109/TNNLS.2023.3293995

Fractal Few-Shot Learning

2024-11-01

0

PRE

AI

F

Fobao Zhou

黄文恺 (Wenkai Huang) *

DOI：10.1109/TNNLS.2023.3293995

原文链接

原文求助

摘要

En 中文

Forming deep feature embeddings is an effective method for few-shot learning (FSL). However, in the case of insufficient samples, overcoming the task complexity while improving the accuracy is still a major challenge. To address this problem, this article considers the consistency between similar data from the fractal perspective, introduces a priori knowledge, and proposes a fractal embedding model by combining FSL with fractal dimension theory for the first time. We improve the original fractal dimension algorithm used to describe image texture roughness to suit a neural network. Moreover, in accordance with the improved algorithm, prior knowledge of the quantized image is integrated into the features to reduce the impact of the data distribution on the model. Experimental results obtained on multiple image benchmark datasets show that the performance of the proposed model exceeds or matches that of previous state-of-the-art models. In addition, the proposed model achieves the best performance in cross-domain scenarios, further illustrating its robustness.

Keyword:

Fractals

Task analysis

Complexity theory

Training

Data models

Adaptation models

Measurement

Deep learning

embedding network

few-shot learning (FSL)

fractal dimension

fractal theory

prior knowledge

Fractal Few-Shot Learning

摘要

期刊

机构

引用论文