Metraipolis › archive › dc_aa4829967bdd

Policy Gradient over History-Dependent Policy Classes for LQR with Domain Randomization — arXiv:2609.16300

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.16300
License (as archived)
CC0-1.0
Archived
2026-10-15T22:39:20.519Z
SHA-256 of the archived text
03e58cc27a20b440e52097268f2c0485f7b6eae4b98e90af1971a35c62ba02c0
Stable URL of this page
https://metraipolis.com/documento/dc_aa4829967bdd

How to cite

Policy Gradient over History-Dependent Policy Classes for LQR with Domain Randomization — arXiv:2609.16300 — arXiv (en), https://arxiv.org/abs/2609.16300 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aa4829967bdd (Metraipolis archive dc_aa4829967bdd, sha256 03e58cc27a20b440e52097268f2c0485f7b6eae4b98e90af1971a35c62ba02c0)
BibTeX
@misc{dcaa4829967bdd,
  title = {Policy Gradient over History-Dependent Policy Classes for LQR with Domain Randomization — arXiv:2609.16300},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.16300},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aa4829967bdd; sha256 03e58cc27a20b440e52097268f2c0485f7b6eae4b98e90af1971a35c62ba02c0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16300. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Domain Randomization (DR) has been widely used to overcome the sim-to-real gap by training a controller on a distribution of simulated environments via reinforcement learning. While DR can achieve robust performance simply using controllers synthesized via policy gradient (PG) methods, the optimization landscape is not well understood, even in the case of linear quadratic regulator (LQR) objectives. To this end, we first study PG of domain randomized LQR over history-dependent policy classes, such as finite impulse response controllers, as they can extend the possibilities of simultaneous stabilization. Second, to find such a stabilizing controller, we propose a curriculum learning based algorithm which gradually expands the memory of the controller. Finally, we show that PG with the proposed algorithm converges globally to the minimizer of a sample average approximation of the DR objective under suitable bounds on the heterogeneity of environments. Empirical results support our findings and highlight promising directions for future work, including nonlinear domain-randomized control.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-16 in eess.SY).
Categorie arXiv: Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2609.16300 — 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 2026-10-15T22:15:42.156Z dalla pagina https://arxiv.org/catchup/eess.SY/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).