{"ID":2870983,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.12031","arxiv_id":"2509.12031","title":"Contractive kinetic Langevin samplers beyond global Lipschitz continuity","abstract":"In this paper, we examine the problem of sampling from log-concave distributions with (possibly) superlinear gradient growth under kinetic (underdamped) Langevin algorithms. Using a carefully tailored taming scheme, we propose two novel discretizations of the kinetic Langevin SDE, and we show that they are both contractive and satisfy a log-Sobolev inequality. Building on this, we establish a series of non-asymptotic bounds in $2$-Wasserstein distance between the law reached by each algorithm and the underlying target measure.","short_abstract":"In this paper, we examine the problem of sampling from log-concave distributions with (possibly) superlinear gradient growth under kinetic (underdamped) Langevin algorithms. Using a carefully tailored taming scheme, we propose two novel discretizations of the kinetic Langevin SDE, and we show that they are both contrac...","url_abs":"https://arxiv.org/abs/2509.12031","url_pdf":"https://arxiv.org/pdf/2509.12031v2","authors":"[\"Iosif Lytras\",\"Panayotis Mertikopoulos\"]","published":"2025-09-15T15:14:45Z","proceeding":"math.PR","tasks":"[\"math.PR\",\"math.NA\",\"stat.ML\"]","methods":"[]","has_code":false}