In this letter, a novel neural architecture search (NAS) method based on reinforcement learning (RL), called RLNAS, is devised to realize the automatic architecture design in the field of hyperspectral unmixing (HU). This method first trains the search network in the constructed self-supervised datasets based on hyperspectral images. The block-based searching and weight-sharing strategies are then introduced to reduce the computational cost in the training phase. The final optimal architecture is obtained by optimizing the multiobjective reward function to balance the trade-off between accuracy and computational efficiency. Compared with the state-of-the-art unmixing algorithms, the proposed RLNAS method can yield better unmixing results on the synthetic and real hyperspectral datasets, which verifies its effectiveness and superiority. In addition, the proposed method offers promising potential of the NAS for HU.

中文翻译：

---

高光谱分解中神经架构搜索的强化学习


在这封信中，设计了一种基于强化学习（RL）的新型神经架构搜索（NAS）方法，称为RLNAS，以实现高光谱分解（HU）领域的自动架构设计。该方法首先在基于高光谱图像构建的自监督数据集中训练搜索网络。然后引入基于块的搜索和权重共享策略以减少训练阶段的计算成本。通过优化多目标奖励函数来平衡准确性和计算效率之间的权衡，得到最终的最优架构。与最先进的解混合算法相比，所提出的 RLNAS 方法可以在合成和真实高光谱数据集上产生更好的解混合结果，验证了其有效性和优越性。此外，所提出的方法为 HU 提供了 NAS 的巨大潜力。