BibTeX

@article{2505.08403v1,
Author        = {Mayank Nautiyal and Andreas Hellander and Prashant Singh},
Title         = {ConDiSim: Conditional Diffusion Models for Simulation Based Inference},
Eprint        = {2505.08403v1},
ArchivePrefix = {arXiv},
PrimaryClass  = {cs.LG},
Abstract      = {We present a conditional diffusion model - ConDiSim, for simulation-based
inference of complex systems with intractable likelihoods. ConDiSim leverages
denoising diffusion probabilistic models to approximate posterior
distributions, consisting of a forward process that adds Gaussian noise to
parameters, and a reverse process learning to denoise, conditioned on observed
data. This approach effectively captures complex dependencies and
multi-modalities within posteriors. ConDiSim is evaluated across ten benchmark
problems and two real-world test problems, where it demonstrates effective
posterior approximation accuracy while maintaining computational efficiency and
stability in model training. ConDiSim offers a robust and extensible framework
for simulation-based inference, particularly suitable for parameter inference
workflows requiring fast inference methods.},
Year          = {2025},
Month         = {May},
Url           = {http://arxiv.org/abs/2505.08403v1},
File          = {2505.08403v1.pdf}
}