Locality Sensitive Hashing in Hyperbolic Space
Published in 42nd International Symposium on Computational Geometry (SoCG 2026), 2026
Locality-sensitive hashing constructions and bounds for approximate nearest neighbor search in hyperbolic space.
Published in 42nd International Symposium on Computational Geometry (SoCG 2026), 2026
Locality-sensitive hashing constructions and bounds for approximate nearest neighbor search in hyperbolic space.
Published in 42nd International Symposium on Computational Geometry (SoCG 2026), 2026
A differentiable GRIL-based topological learning layer for end-to-end graph and bio-activity prediction.
Published in 39th Conference on Neural Information Processing Systems (NeurIPS 2025), 2025
Extending the Johnson-Lindenstrauss lemma to non-Euclidean geometry with fine-grained error analysis based on geometric deviation.
Published in 42nd International Conference on Machine Learning (ICML), 2025
A topologically interpretable GNN with a novel topological discrepancy loss is proved to be uniquely optimized by ground truth.
Published in The Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024), 2024
Generalizing MDS to accommodate non-Euclidean and non-metric inputs via symmetric bilinear forms.
Published in Transactions on Machine Learning Research (TMLR), 2024
A symmetry-breaking approach that improves higher-order link prediction expressivity for hypergraph representation learning.
Published in 41st International Conference on Machine Learning (ICML), 2024
Algorithms for optimally fixing classifiers in a cooperative network to maximize aggregate or egalitarian accuracy.
Published in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Presenting a massive 10K real-world scene benchmark dataset for 3D vision with 51.2 million frames.
Published in 2nd Annual Workshop on Topology, Algebra, and Geometry in Machine Learning (TAG-ML, ICML 2023 Workshop), 2023
A stable and differentiable 2-parameter persistence vectorization for machine learning; oral presentation at the ICML 2023 TAG-ML workshop.
Published in Journal of Applied and Computational Topology (JACT), 2022
Generalize the persistence algorithm to compute decompositions of multi-parameter persistence modules.
Published in 34th International Symposium on Computational Geometry (SoCG), 2018
Present an efficient algorithm to compute the bottleneck distance 2-parameter interval decomposable models.