Sharper Bounds for Chebyshev Moment Matching, with Applications

ArXi:2408.12385v3 Announce Type: replace-cross We study the problem of approximately recovering a probability distribution given noisy measurements of its Chebyshe polynomial moments. This problem arises broadly across algorithms, statistics, and machine learning. By leveraging a global decay bound on the coefficients in the Chebyshe expansion of any Lipschitz function, we sharpen prior work, proving that accurate recovery in the Wasserstein distance is possible with noise than previously known.