Quantum Algorithms for Gibbs Expectation of Non-log-concave and Heavy-tailed Distributions

Name: Quantum Algorithms for Gibbs Expectation of Non-log-concave and Heavy-tailed Distributions
Uploaded: 2026-10-21T18:42:38.122161+00:00
Duration: 809 s
Description: The research develops a quantum framework to estimate Gibbs expectations from complex, non-log-concave distributions and heavy-tailed data. The method introduces an unbiased estimator via QA-MLMC and uses a change-of-measure approach based on dissipativity. This yields a quadratic speedup over classical methods, achieving $\mathcal{O}(\epsilon^{-1})$ complexity for accuracy $\epsilon$.

The research develops a quantum framework to estimate Gibbs expectations from complex, non-log-concave distributions and heavy-tailed data. The method introduces an unbiased estimator via QA-MLMC and uses a change-of-measure approach based on dissipativity. This yields a quadratic speedup over classical methods, achieving $\mathcal{O}(\epsilon^{-1})$ complexity for accuracy $\epsilon$.

Open the full episode