A0293
Title: Cluster quilting: Spectral clustering for patchwork learning
Authors: Lili Zheng - University of Illinois Urbana-Champaign (United States) [presenting]
Andersen Chang - Baylor College of Medicine (United States)
Genevera Allen - Rice University (United States)
Abstract: Patchwork learning arises as a new and challenging data collection paradigm where both samples and features are observed in fragmented subsets. Due to technological limits, measurement expense, or multimodal data integration, such patchwork data structures are frequently seen in neuroscience, healthcare, and genomics, among others. Instead of analyzing each data patch separately, it is highly desirable to extract comprehensive knowledge from the whole data set. The focus is on the clustering problem in patchwork learning, aiming at discovering clusters amongst all samples even when some are never jointly observed for any feature. A novel spectral clustering method is proposed called cluster quilting, consisting of (i) Patch ordering that exploits the overlapping structure amongst all patches, (ii) Patchwise SVD, (iii) Sequential linear mapping of top singular vectors for patch overlaps, followed by (iv) K-means on the combined and weighted singular vectors. Under a sub-Gaussian mixture model, theoretical guarantees are established via a non-asymptotic misclustering rate bound that reflects both properties of the patch-wise observation regime as well as the clustering signal and noise dependencies. The cluster quilting algorithm is also validated through extensive empirical studies on both simulated and real data sets in neuroscience and genomics, where it discovers more accurate and scientifically more plausible clusters than other approaches.
