Ruqi Zhang
@ruqi-zhang.bsky.social
Assistant Professor@PurdueCS | PhD@Cornell | Probabilistic machine learning, Bayesian deep learning, Sampling, MCMC
Cyclical MCMC is a popular method for sampling from complex multimodal distributions. In our #NeurIPS2024 paper, we extend this approach to discrete distributions with novel designs tailored for discrete variables. Applications include graphic models, energy-based models, and language models.
December 13, 2024 at 9:01 AM
Cyclical MCMC is a popular method for sampling from complex multimodal distributions. In our #NeurIPS2024 paper, we extend this approach to discrete distributions with novel designs tailored for discrete variables. Applications include graphic models, energy-based models, and language models.