{"id":"arxiv/2601.02797","url":"https://arxiv.org/abs/2601.02797","tags":["math.OC"],"score":0,"title":"Log-Polynomial Optimization","domain":"math.OC","source":"arxiv","authors":["Jiyoung Choi","Jiawang Nie","Xindong Tang","Suhan Zhong"],"reviews":0,"subject":"math.OC","summary":"We study an optimization problem in which the objective is given as a sum of logarithmic-polynomial functions. This formulation is motivated by statistical estimation principles such as maximum likelihood estimation, and by loss functions including cross-entropy and Kullback-Leibler divergence. We propose a hierarchy of moment relaxations based on the truncated $K$-moment problems to solve log-polynomial optimization. We provide sufficient conditions for the hierarchy to be tight and introduce a numerical method to extract the global optimizers when the tightness is achieved. In addition, we…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-01-06","reactions":{}}