Generalizing Metropolis-Hastings for DAG Sampling

The paper introduces duet sampling, a generalization of the Metropolis-Hastings algorithm that selects two proposal distributions from a pool at each iteration to resample parents for multiple nodes in a directed acyclic graph (DAG). This approach enables Markov chain moves that mix more efficiently without extra computational cost, outperforming existing state‑of‑the‑art methods in experiments. The authors demonstrate that their novel moves, both alone and combined, achieve superior performance even when previous methods also adopt duet sampling.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.