Research map / Learning theory and algorithms / Prediction and decision losses
Stable Vectorization of Multiparameter Persistent Homology using Signed Barcodes as Measures
David Loiseaux, Luis Scoccola, Mathieu Carrière, Magnus Bakke Botnan, Steve Oudot
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 |
|---|---|---|---|
| Use Hilbert decomposition signed barcode instead of full barcode (one-parameter) | barcode vectorizations (PI, PL, SWK) · one-parameter point cloud and graph datasets | accuracy | No clear effect |
| Use Hilbert signed barcode instead of pervec histogram | PV · one-parameter datasets | accuracy | Improved |
| MP-HSM-C vs other multiparameter descriptors (MP-L, MP-I, MP-K, MP-H) | multiparameter persistence baselines · UCR time series | accuracy | Improved |
| MP-HSM-C vs non-topological baselines (Euclidean NN, DTW) | non-topological baselines · UCR time series | accuracy | No clear effect |
| Sliced Wasserstein vectorization (MP-HSM-SW) on alpha complex filtered by DTM | MP-HSM-C · UCR time series | accuracy | Hurt |
| Filter a Rips complex by density instead of an alpha complex with DTM for MP-HSM-SW (ablation) | MP-HSM-SW · UCR time series | accuracy | Improved |
| MP-HSM-C vs multiparameter topological baselines incl. GRIL | topological baselines · graph classification | accuracy | Improved |
| MP-ESM-C vs unsupervised baselines | unsupervised virtual screening baselines · Cleves-Jain | enrichment factor | Improved |
| Signed barcode vectorizations vs raw Hilbert and Euler functions | MP-H and MP-E · Cleves-Jain | enrichment factor | Improved |
Closest papers by meaning
- Differentiability and Optimization of Multiparameter Persistent Homology (ICML 2024 · similarity 0.94)
- A Framework for Fast and Stable Representations of Multiparameter Persistent Homology Decompositions (NeurIPS 2023 · similarity 0.92)
- Graphcode: Learning from multiparameter persistent homology using graph neural networks (NeurIPS 2024 · similarity 0.90)
- Persistent Homology for High-dimensional Data Based on Spectral Methods (NeurIPS 2024 · similarity 0.87)
- The Flood Complex: Large-Scale Persistent Homology on Millions of Points (NeurIPS 2025 · similarity 0.86)
- On topological descriptors for graph products (NeurIPS 2025 · similarity 0.86)
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.
