Research output per year
Research output per year
Description
This dataset contains four real-world sub-datasets with data embedded into Poincare ball models, including Olsson's single-cell RNA expression data, CIFAR10, Fashion-MNIST and mini-ImageNet. Each sub-dataset has two corresponding files: one is the data file, the other one is the pre-computed reference points for each class in the sub-dataset. Please refer to our paper (https://arxiv.org/pdf/2109.03781.pdf) and codes (https://github.com/thupchnsky/PoincareLinearClassification) for more details.
| Date made available | Jun 1 2023 |
|---|---|
| Publisher | University of Illinois Urbana-Champaign |
Keywords
- Hyperbolic space
- Machine learning
- Support vector machine
- Poincare ball models
- Perceptron algorithm
Research output
- 1 Conference contribution
-
Highly Scalable and Provably Accurate Classification in Poincaré Balls
Chien, E., Pan, C., Tabaghi, P. & Milenkovic, O., 2021, Proceedings - 21st IEEE International Conference on Data Mining, ICDM 2021. Bailey, J., Miettinen, P., Koh, Y. S., Tao, D. & Wu, X. (eds.). Institute of Electrical and Electronics Engineers Inc., p. 61-70 10 p. (Proceedings - IEEE International Conference on Data Mining, ICDM; vol. 2021-December).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
Open Access