Research map / Probabilistic and scientific modeling / Variational and generative sampling
MPCC: Matching Priors and Conditionals for Clustering
Nicolás Astorga, Pablo Huijse, Pavlos Protopapas, Pablo Estévez
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 |
|---|---|---|---|
| increase encoder updates E_steps from 1 to 4 (ablation) | MPCC · CIFAR-10 | FID: 11.38 -> 7.50 | Improved |
| increase encoder updates E_steps from 1 to 4 (ablation) | MPCC · CIFAR-10 | IS: 8.82 -> 9.25 | Improved |
| increase encoder updates E_steps from 1 to 4 (ablation) | MPCC · CIFAR-10 | MSE (x1e-3): 1.34 -> 0.331 | Improved |
| increase encoder updates E_steps from 1 to 4 (ablation) | MPCC · CIFAR-10 | clustering accuracy (%): 41.31 -> 38.27 | No clear effect |
| increase prior learning rate eta_p from 2e-4 to 6e-4 (ablation) | MPCC · CIFAR-10 | clustering accuracy (%): 38.27 -> 61.99 | Improved |
| share parameters between encoder and discriminator (ablation) | AIM-MPCC · CIFAR-10 | FID: 21.55 -> 10.42 | Improved |
| share parameters between encoder and discriminator (ablation) | AIM-MPCC · CIFAR-10 | IS: 8.24 -> 9.09 | Improved |
| add learnable GMM prior instead of standard Normal prior (ablation) | MPCC (S) · CIFAR-10 | FID: 10.42 -> 6.59 | Improved |
| add learnable GMM prior instead of standard Normal prior (ablation) | MPCC (S) · CIFAR-10 | IS: 9.09 -> 9.49 | Improved |
| use class-conditional generation instead of unconditional | MPCC · CIFAR-10 | FID: 6.59 -> 5.69 | Improved |
| use batch size larger than 50 (ablation) | MPCC · CIFAR-10 | clustering accuracy | Hurt |
| use D_steps=4 rather than smaller values (ablation) | MPCC · CIFAR-10 | mode collapse | Improved |
Closest papers by meaning
- Top-Down Deep Clustering with Multi-Generator GANs (AAAI 2022 · similarity 0.88)
- ClusterGAN: Latent Space Clustering in Generative Adversarial Networks (AAAI 2019 · similarity 0.88)
- Nearest Neighbor Matching for Deep Clustering (CVPR 2021 · similarity 0.87)
- Samples Are Not Equal: A Sample Selection Approach for Deep Clustering (ICLR 2026 · similarity 0.86)
- From Logits to Hierarchies: Hierarchical Clustering made Simple (ICML 2025 · similarity 0.84)
- You Can Trust Your Clustering Model: A Parameter-free Self-Boosting Plug-in for Deep Clustering (NeurIPS 2025 · similarity 0.84)
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.
