Vol. 3 No. 9 (2024)
Articles

Improving Few-Shot Learning via Dual-Phase Manifold Embedding Optimization

Published 2024-12-30

How to Cite

Zhang, Z., & Radcliffe, E. (2024). Improving Few-Shot Learning via Dual-Phase Manifold Embedding Optimization. Journal of Computer Technology and Software, 3(9). Retrieved from https://ashpress.org/index.php/jcts/article/view/107

Abstract

The pursuit of Artificial General Intelligence (AGI) necessitates models capable of rapid adaptation to novel tasks with minimal data, akin to human learning. This paper introduces TSMB (TWO-STAGE MANIFOLD-BASED FEW-SHOT LEARNING), a novel approach to Few-Shot Learning (FSL) that leverages unsupervised learning to harness the geometric distribution of data across tasks. TSMB refines feature representations through a two-stage process: first, by leveraging the topological structure of high-dimensional data to fine-tune general features, and second, by generating virtual samples to integrate semi-supervised learning. This method aims to address the challenges of data scarcity and overfitting, common in FSL. The model comprises a backbone network for feature extraction, a manifold learning module to capture topological results, a manifold support point section to assist learning, and a denoising prototype classifier for decision-making. TSMB demonstrates enhanced generality and performance, offering a promising direction for advancing FSL towards more ethical, sustainable, and effective AI applications.