Dimensionality Reduction Meets Network Science: Sensemaking on UMAP’s kNN Graph

Researchers from Georgia Tech and Apple have shown that applying standard graph algorithms to UMAP's internal k-nearest-neighbor (kNN) graph can enhance data sensemaking, outperforming or complementing purpose-built methods. In a paper published July 2026, Duen Horng (Polo) Chau, Donghao Ren, Fred Hohman, and Dominik Moritz demonstrate that PageRank, k-core decomposition, and clustering coefficient on the kNN graph improve tasks like exemplar selection and density-based clustering on MNIST and Fashion MNIST datasets.

content type paper /research/ published July 2026 Dimensionality Reduction Meets Network Science: Sensemaking on UMAP’s kNN Graph AuthorsDuen Horng Polo Chau, Donghao Ren, Fred Hohman, Dominik Moritz Dimensionality Reduction Meets Network Science: Sensemaking on UMAP’s kNN Graph AuthorsDuen Horng Polo Chau, Donghao Ren, Fred Hohman, Dominik Moritz While UMAP is widely used for exploring high-dimensional data, typical workflows focus on its lower-dimensional embedding, largely overlooking the rich k-nearest-neighbor kNN graph that UMAP constructs internally. This graph encodes the data manifold in its original high-dimensional space, before the distortion that UMAP’s 2D projection introduces. We demonstrate the untapped potential of this internal representation, showing how standard graph algorithms applied to this graph enhance data sensemaking: 1 PageRank identifies representative data points, 2 k-core decomposition reveals dense core regions versus sparse periphery, and 3 clustering coefficient detects tight-knit neighborhoods with highly-similar data points. Through quantitative and qualitative evaluation on MNIST and Fashion MNIST, we show that these graph-based analyses are not only practical but also competitive with or complementary to purpose-built methods e.g., k-medoids for exemplar selection, HDBSCAN for density-based clustering . Generative and Contrastive Graph Representation Learning September 29, 2025 research area Methods and Algorithms /research/?domain=Methods%20and%20Algorithms Self-supervised learning SSL on graphs generates node and graph representations i.e., embeddings that can be used for downstream tasks such as node classification, node clustering, and link prediction. Graph SSL is particularly useful in scenarios with limited or no labeled data. Existing SSL methods predominantly follow contrastive or generative paradigms, each excelling in different tasks: contrastive methods typically perform well on… Regularized Training of Nearest Neighbor Language Models August 17, 2022 research area Speech and Natural Language Processing /research/?domain=Speech%20and%20Natural%20Language%20Processing Workshop at NAACL /research/?event=NAACL%20Workshop Including memory banks in a natural language processing architecture increases model capacity by equipping it with additional data at inference time. In this paper, we build upon kNN-LM, which uses a pre-trained language model together with an exhaustive kNN search through the training data memory bank to achieve state-of-the-art results. We investigate whether we can improve the kNN-LM performance by instead training a LM with the knowledge…