BlockEcho: Retaining Long-Range Dependencies for Imputing Block-Wise Missing Data
This addresses a critical issue in data imputation for real-world applications where block-wise gaps degrade machine learning tasks, offering a novel solution with broad impact.
The paper tackles the problem of imputing block-wise missing data, which is more challenging than scattered missing data, by proposing BlockEcho, a method that integrates Matrix Factorization with Generative Adversarial Networks to retain long-range dependencies, achieving superior performance over state-of-the-art methods, especially at high missing rates.
Block-wise missing data poses significant challenges in real-world data imputation tasks. Compared to scattered missing data, block-wise gaps exacerbate adverse effects on subsequent analytic and machine learning tasks, as the lack of local neighboring elements significantly reduces the interpolation capability and predictive power. However, this issue has not received adequate attention. Most SOTA matrix completion methods appeared less effective, primarily due to overreliance on neighboring elements for predictions. We systematically analyze the issue and propose a novel matrix completion method ``BlockEcho" for a more comprehensive solution. This method creatively integrates Matrix Factorization (MF) within Generative Adversarial Networks (GAN) to explicitly retain long-distance inter-element relationships in the original matrix. Besides, we incorporate an additional discriminator for GAN, comparing the generator's intermediate progress with pre-trained MF results to constrain high-order feature distributions. Subsequently, we evaluate BlockEcho on public datasets across three domains. Results demonstrate superior performance over both traditional and SOTA methods when imputing block-wise missing data, especially at higher missing rates. The advantage also holds for scattered missing data at high missing rates. We also contribute on the analyses in providing theoretical justification on the optimality and convergence of fusing MF and GAN for missing block data.