Research map / Probabilistic and scientific modeling / Variational and generative sampling
Topological Autoencoders
Michael Moor, Max Horn, Bastian Rieck, Karsten Borgwardt
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 |
|---|---|---|---|
| Add the topological loss to a standard autoencoder | AE · Spheres | DKL_0.1: 0.746 -> 0.326 | Improved |
| Add the topological loss to a standard autoencoder | AE · Fashion-MNIST | DKL_0.1: 0.068 -> 0.054 | Improved |
| Add the topological loss to a standard autoencoder | AE · MNIST | DKL_0.1: 0.155 -> 0.110 | Improved |
| Add the topological loss to a standard autoencoder | AE · CIFAR-10 | DKL_0.1: 0.035 -> 0.019 | Improved |
| Add the topological loss to a standard autoencoder | AE · Spheres | Data MSE: 0.8155 -> 0.8681 | Hurt |
| Add the topological loss to a standard autoencoder | AE · Fashion-MNIST | Data MSE: 0.1020 -> 0.1207 | Hurt |
| Add the topological loss to a standard autoencoder | AE · MNIST | Data MSE: 0.1373 -> 0.1388 | Hurt |
| Add the topological loss to a standard autoencoder | AE · CIFAR-10 | Data MSE: 0.1403 -> 0.1398 | Improved |
| Use 1-dimensional persistence features rather than only 0-dimensional features (ablation) | TopoAE · Experiments | runtime | Hurt |
| Add the topological loss to a standard autoencoder | AE · Spheres | W_1: 3.99±0.037 -> 3.73±0.076 | Improved |
| Add the topological loss to a standard autoencoder | AE · Spheres | W_2: 0.469±0.0053 -> 0.459±0.0055 | Improved |
| Add the topological loss to a standard autoencoder | AE · Spheres | W_infinity: 0.154±0.00128 -> 0.152±0.00268 | Improved |
Closest papers by meaning
- Connectivity-Optimized Representation Learning via Persistent Homology (ICML 2019 · similarity 0.90)
- Towards Scalable Topological Regularizers (ICLR 2025 · similarity 0.87)
- Graph Geometry-Preserving Autoencoders (ICML 2024 · similarity 0.87)
- Diffeomorphic interpolation for efficient persistence-based topological optimization (NeurIPS 2024 · similarity 0.86)
- Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity (ICLR 2025 · similarity 0.85)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.85)
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.
