Metraipolis › archive › dc_f99303d451d3

Self-Normalized Inference for Constant-Stepsize Temporal-Difference Learning under Markovian Sampling — arXiv:2608.10896

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10896
License (as archived)
CC0-1.0
Archived
2026-10-19T06:19:51.257Z
SHA-256 of the archived text
fc58d3497afb89dcefe25c341c1fdec0274b569197f2536885912bedf8bced83
Stable URL of this page
https://metraipolis.com/documento/dc_f99303d451d3

How to cite

Self-Normalized Inference for Constant-Stepsize Temporal-Difference Learning under Markovian Sampling — arXiv:2608.10896 — arXiv (en), https://arxiv.org/abs/2608.10896 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f99303d451d3 (Metraipolis archive dc_f99303d451d3, sha256 fc58d3497afb89dcefe25c341c1fdec0274b569197f2536885912bedf8bced83)
BibTeX
@misc{dcf99303d451d3,
  title = {Self-Normalized Inference for Constant-Stepsize Temporal-Difference Learning under Markovian Sampling — arXiv:2608.10896},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10896},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f99303d451d3; sha256 fc58d3497afb89dcefe25c341c1fdec0274b569197f2536885912bedf8bced83},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10896. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Constant-stepsize temporal-difference (TD) learning is attractive for policy evaluation, but inference from a single Markov trajectory must account for serial dependence and a stepsize-dependent stationary target. For fixed-stepsize linear TD, we establish a functional central limit theorem whose covariance retains the multiplicative component induced by the random TD matrix and the stationary iterate error. We then derive a joint functional limit for parallel Richardson--Romberg (RR) recursions driven by the same trajectory. A Brownian-bridge self-normalizer yields asymptotically pivotal confidence regions for prespecified state-value contrasts without estimating the long-run covariance or selecting a bandwidth or batch length. For such a contrast, the procedure admits a one-pass implementation whose memory does not grow with the trajectory length. At a fixed stepsize, the inferential center is the RR stationary target. We also study horizon-indexed designs in which the stepsize remains constant within each run and decreases across longer horizons. Under an explicit RR-dependent rate window, the residual RR target shift, multiplicative remainder, and initialization effect are negligible at the root-$n$ scale, yielding inference for the projected Bellman solution. Experiments on FrozenLake and Garnet illustrate stationary-target coverage, RR target correction, and the finite-sample behavior of the horizon-indexed design.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.LG).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.10896 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T06:11:45.149Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).