Research map / Image generation / GANs
GATSBI: Generative Adversarial Training for Simulation-Based Inference
Poornima Ramesh, Jan-Matthis Lueckmann, Jan Boelts, Álvaro Tejero-Cantero, David S. Greenberg, Pedro J. Goncalves, Jakob H. Macke
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 GATSBI rather than NLE | Shallow water model · Shallow water model | correlation of posterior mean with ground-truth parameters: 0.88 \pm 0.10 | Improved |
| Tune the GAN architecture and training hyperparameters specifically for the Two-Moons model (ablation) | GATSBI · Two-Moons | C2ST score | Improved |
| Increase the simulation budget | GATSBI · SLCP | C2ST score | Improved |
| Increase the simulation budget | GATSBI · Two-Moons | C2ST score | No clear effect |
| Use sequential inference instead of amortised GATSBI | GATSBI · SBI benchmark problems | C2ST score | Improved |
| Use sequential GATSBI instead of amortised GATSBI | GATSBI · SBI benchmark problems | C2ST score relative to flow-based methods | Hurt |
| Use GATSBI instead of NPE | Shallow water model · Shallow water model | posterior calibration | No clear effect |
| Use GATSBI instead of NPE | Noisy camera model · Noisy camera model | coherence of posterior samples | Improved |
| Use GATSBI instead of NLE | Shallow water model · Shallow water model | recovery of depth-profile shape | Improved |
Closest papers by meaning
- Generative Bayesian Inference with GANs (JMLR 2026 · similarity 0.89)
- Generalized Adversarially Learned Inference (AAAI 2021 · similarity 0.86)
- Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models (NeurIPS 2025 · similarity 0.85)
- Bayesian GAN (NeurIPS 2017 · similarity 0.85)
- Implicit Variational Inference for High-Dimensional Posteriors (NeurIPS 2023 · similarity 0.85)
- FNOPE: Simulation-based inference on function spaces with Fourier Neural Operators (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.
