提出新型深度展开网络，实现块稀疏信号高效恢复。
Deep Unfolded Latent Optimally Partitioned-l2/l1 Networks for Data-driven Block-Sparse Recovery
- 采用隐式微分与权重分解技术提升稳定性
- 在未知分组下仍保持良好恢复性能
- 适合需要抗脉冲噪声的信号处理场景
凸的潜在最优分组（LOP）-l2/l1方法可实现未知分组下的块稀疏信号恢复，但依赖人工调参，且其近端算子求导存在数值不稳定性，阻碍了通过深度展开（DU）实现自动调参。为此，本文提出两种架构：一种基于隐式微分的稳定框架，另一种利用深度权重分解（DWF）的灵活变体，后者还支持非凸平滑数据保真项。数值实验表明，DU-LOP-l2/l1方法具有竞争力的恢复性能，并对脉冲噪声具有强鲁棒性。
原文摘要 · Abstract (English)
The convex Latent Optimal Partition (LOP)-l2/l1 approach enables block-sparse signal recovery with unknown partitions but relies on manual hyperparameter tuning. Additionally, numerical instability in differentiating its proximal operator prevents its automatic parameter tuning via Deep Unfolding (DU). To address these limitations, we propose two architectures: a stable framework utilizing implicit differentiation and a flexible variant leveraging Deep Weight Factorization (DWF). The DWF-based approach also supports nonconvex smooth data fidelity terms. Numerical experiments demonstrate that DU-LOP-l2/l1 yields competitive performance and high resilience against impulsive noise.
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。