arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2607.07083eess.SYcs.LGcs.SY
收藏

用于主动概率子采样的先验感知和上下文引导分组采样

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Department of Artificial Intelligence, Korea University, Seoul, Republic of Korea(人工智能系,韩国大学,首尔,大韩民国)

Beomgu Kang, Hyunseok Seo

AI总结:

研究针对主动概率子采样方法不足,提出先验感知和上下文引导分组主动DPS(PGA-DPS),整合先验信息采样模式与基于组的top-k采样,经理论分析和实证验证,在多任务中表现优于其他方法。

摘要:

子采样显著减少测量数量,简化数据处理和传输开销,缩短采集时间。最近引入的主动深度概率子采样(A-DPS)联合优化子采样模式和下游任务模型。但该方法未充分利用数据集先验,依赖top-1采样阻碍优化。本文通过整合源自训练数据集的确定性先验信息采样模式及基于组的top-k采样增强A-DPS,提出先验感知和上下文引导分组主动DPS(PGA-DPS)。还提供理论分析并通过实证结果验证。在分类、图像重建和分割三个任务上评估,PGA-DPS优于A-DPS、DPS和其他采样方法。

英文摘要:

Subsampling significantly reduces the number of measurements, thereby streamlining data processing and transfer overhead, and shortening acquisition time across diverse real-world applications. The recently introduced Active Deep Probabilistic Subsampling (A-DPS) approach jointly optimizes both the subsampling pattern and the downstream task model, enabling instance- and subject-specific sampling trajectories and effective adaptation to new data at inference time. However, this approach does not fully leverage valuable dataset priors and relies on top-1 sampling, which can impede the optimization process. Herein, we enhance A-DPS by integrating a deterministic (fixed) prior-informed sampling pattern derived from the training dataset, along with group-based sampling via top-k sampling, to achieve more robust optimization, method we call Prior-aware and context-guided Group-based Active DPS (PGA-DPS). We also provide a theoretical analysis supporting improved optimization via group sampling, and validate this with empirical results. We evaluated PGA-DPS on three tasks: classification, image reconstruction, and segmentation, using the MNIST, CIFAR-10, fastMRI knee, and hyperspectral AeroRIT datasets, respectively. In every case, PGA-DPS outperformed A-DPS, DPS, and all other sampling methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑