Research map / Probabilistic and scientific modeling / Variational and generative sampling
Cooperative Learning of Energy-Based Model and Latent Variable Model via MCMC Teaching
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| train descriptor and generator cooperatively instead of generator alone (Algorithm G) (ablation) | CoopNets · ImageNet 10 scenes, n=1100 | Inception score: 2.61 -> 3.79 | Helped |
| train descriptor and generator cooperatively instead of descriptor alone (persistent CD) (ablation) | CoopNets · ImageNet 10 scenes, n=1100 | Inception score: 1.51 -> 3.79 | Helped |
| cooperative learning versus DCGAN with small training set | CoopNets · ImageNet 10 scenes, n=50 | Inception score: 2.26 -> 2.66 | Helped |
| cooperative learning versus DCGAN at large n | CoopNets · ImageNet 10 scenes, n=1100 | Inception score: 3.47 -> 3.79 | Helped |
| use CoopNets generator for occlusion recovery instead of DCGAN generator | CoopNets · CelebA M30 | recovery error: 0.211 -> 0.115 | Helped |
| use CoopNets instead of interpolation baseline inter-1 | CoopNets · CelebA M30 | recovery error: 0.120 -> 0.115 | Helped |
| use CoopNets instead of interpolation baseline inter-4 | CoopNets · CelebA M50 | PSNR: 14.164 -> 15.105 | Helped |
| cooperative learning versus DCGAN and VAE | CoopNets · single ImageNet category | softmax class probability, top-5 error, structural similarity | Helped |
Closest papers by meaning
- Unbiased Contrastive Divergence Algorithm for Training Energy-Based Latent Variable Models (ICLR 2020 · similarity 0.88)
- Learning Non-Convergent Non-Persistent Short-Run MCMC Toward Energy-Based Model (NeurIPS 2019 · similarity 0.88)
- On the Anatomy of MCMC-Based Maximum Likelihood Learning of Energy-Based Models (AAAI 2020 · similarity 0.88)
- Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation (AAAI 2021 · similarity 0.87)
- Learning Energy-Based Models by Cooperative Diffusion Recovery Likelihood (ICLR 2024 · similarity 0.85)
- Joint Learning of Energy-based Models and their Partition Function (ICML 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.