Research map / Learning theory and algorithms / Matrix and tensor methods
Quantifying the Informativeness of Similarity Measurements
Austin J. Brockmeier, Tingting Mu, Sophia Ananiadou, John Y. Goulermas
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| select dimensionality by Bures informativeness instead of using original distances | 1-NN classification on kernel embedding · ionosphere (UCI) | accuracy (%): 85 -> 94 | Improved |
| select dimensionality by Bures informativeness instead of using original distances | 1-NN classification · breast cancer, sonar, iris, glass, ecoli (UCI) | accuracy (%): 95 vs 95; 83 vs 83; 94 vs 94; 68 vs 68; 80 vs 80 | No clear effect |
| use Bures informativeness instead of supervised CKA1 dimension selection | kernel dimensionality reduction · 7 UCI data sets | average loss margin: 7.7 -> 1.2 | Improved |
| use Bures informativeness instead of Euclidean informativeness | kernel dimensionality reduction · 7 UCI data sets | average loss margin: 10.6 -> 1.2 | Improved |
| use Bures informativeness instead of cosine informativeness | kernel dimensionality reduction · 7 UCI data sets | average loss margin: 3.2 -> 1.2 | Improved |
| use HSIC-based informativeness instead of Bures | kernel dimensionality reduction · 7 UCI data sets | average difference vs original: 9.5 vs -1.1 | Hurt |
| use Bartlett or Lawley test statistics instead of informativeness | dimensionality selection · 7 UCI data sets | average difference vs original: 17.3 vs -1.1 | Hurt |
| use centered kernel alignment (CKA) informativeness instead of Bures | dimensionality selection · 7 UCI data sets | average difference vs original: 27.2 vs -1.1 | Hurt |
| use cosine informativeness instead of supervised CKA dimension selection | kernel dimensionality reduction · 7 UCI data sets | average difference vs original: 0.9 vs 5.5 | Improved |
Closest papers by meaning
- Robust and Fast Measure of Information via Low-Rank Representation (AAAI 2023 · similarity 0.85)
- Sample Complexity of Correlation Detection in the Gaussian Wigner Model (ICML 2025 · similarity 0.84)
- Robust Spectral Clustering with Rank Statistics (JMLR 2024 · similarity 0.84)
- Integrating Tensor Similarity to Enhance Clustering Performance (TPAMI 2020 · similarity 0.84)
- How to Capture Higher-order Correlations? Generalizing Matrix Softmax Attention to Kronecker Computation (ICLR 2024 · similarity 0.83)
- Spectral Analysis of Representational Similarity with Limited Neurons (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
