自适应变换框架提升模型表示能力，无需先验知识即可自动选择最优变换。
General Transform: A Unified Framework for Adaptive Transform to Enhance Representations
- 基于数据学习动态映射，自动适配不同任务与数据集
- 在视觉与语言任务中均超越传统变换方法
- 适合缺乏领域知识但需高效特征提取的场景
离散变换（如离散傅里叶变换）在机器学习中被广泛用于提取有意义特征以提升模型性能。然而，由于可用变换种类繁多，选择合适的变换通常依赖对数据集特性的理解，当此类知识不可得时，该方法效果受限。本文提出通用变换（General Transform, GT），一种面向机器学习应用的自适应变换表示框架。与传统变换不同，GT 能够根据目标数据集和任务学习数据驱动的映射。实验表明，引入 GT 的模型在计算机视觉与自然语言处理任务中均优于传统变换方法，验证了其在多种学习场景下的有效性。
原文摘要 · Abstract (English)
Discrete transforms, such as the discrete Fourier transform, are widely used in machine learning to improve model performance by extracting meaningful features. However, with numerous transforms available, selecting an appropriate one often depends on understanding the dataset's properties, making the approach less effective when such knowledge is unavailable. In this work, we propose General Transform (GT), an adaptive transform-based representation designed for machine learning applications. Unlike conventional transforms, GT learns data-driven mapping tailored to the dataset and task of interest. Here, we demonstrate that models incorporating GT outperform conventional transform-based approaches across computer vision and natural language processing tasks, highlighting its effectiveness in diverse learning scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。