资源感知的参数高效模型适配用于星载高维数据
Resource-Aware Parameter-Efficient Model Adaptation for Onboard High-Dimensional Data
浏览论文内容
中文总结 AI 辅助
提出NE-LoRA参数高效适配框架，结合低秩与非线性分支及差异化训练策略，在带宽受限的星载高光谱场景中以少量参数更新实现优于LoRA、媲美全量微调的性能。
中文摘要 AI 辅助
星载卫星模型经常需要更新，但适应早期数据分布的权重可能很快过时。然而，在轨更新大规模模型参数面临重大挑战，因为低地球轨道（LEO）卫星系统的上行链路带宽有限，特别是对于高光谱卫星图像，高维光谱-空间输入导致模型规模和更新成本增加。现有的全量微调方法因此重新训练成本高昂，且在严格的通信约束下难以部署。为解决这一挑战，我们提出了NE-LoRA，一种用于带宽受限的星载高光谱模型更新的参数高效适配框架。NE-LoRA结合了一个主低秩分支和一个非线性辅助分支，以捕获全局更新趋势和复杂的光谱-空间变化。此外，我们引入了一种针对多矩阵适配器的差异化训练策略，其动机来自不同适配器矩阵的不对称初始化和梯度动态。在四个高光谱数据集和三个代表性骨干模型上的实验表明，NE-LoRA始终优于基于LoRA的基线，并且与全量微调相比具有竞争力，在某些情况下甚至优于全量微调。在评估的设置中，NE-LoRA平均仅更新总参数的一小部分，同时保持低部署开销，为星载高光谱适配提供了有利的精度-通信权衡。
英文摘要
Onboard satellite models often require frequent updates, but the weights adapted to earlier data distributions can quickly become outdated. However, updating large-scale model parameters in orbit presents significant challenges due to the limited uplink bandwidth of Low Earth Orbit (LEO) satellite systems, particularly for hyperspectral satellite imagery, where high-dimensional spectral-spatial inputs lead to increased model size and update costs. Existing full fine-tuning methods are thus expensive to retrain and difficult to deploy under strict communication constraints. To address this challenge, we propose NE-LoRA, a parameter-efficient adaptation framework for bandwidth-constrained onboard hyperspectral model updates. NE-LoRA combines a primary low-rank branch with a nonlinear auxiliary branch to capture both global update trends and complex spectral-spatial variations. Additionally, we introduce a differentiated training strategy for multi-matrix adapters, motivated by the asymmetric initialization and gradient dynamics of different adapter matrices. Experiments on four hyperspectral datasets and three representative backbone models demonstrate that NE-LoRA consistently outperforms LoRA-based baselines and remains competitive with, and in several cases superior to, full fine-tuning. Across the evaluated settings, NE-LoRA updates only a small fraction of the total parameters on average while preserving low deployment overhead, offering a favorable accuracy-communication trade-off for onboard hyperspectral adaptation.
发表机构
- Beijing University of Posts and Telecommunications（北京邮电大学）
- Wuhan University（武汉大学）
- Communication University of China（中国传媒大学）
- University of Luxembourg（卢森堡大学）
机构由 AI 辅助整理，请以论文原文为准。
