Metraipolis › archive › dc_f485d183f0a2
DISEIL: Demonstration Distillation for Sample-Efficient Imitation Learning — arXiv:2609.08123
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.08123
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T13:59:26.487Z
- SHA-256 of the archived text
759eb4a5f0eb1b6635bef980457357b1f4d71da0c4ac0f30ada7744d0b44260c- Stable URL of this page
https://metraipolis.com/documento/dc_f485d183f0a2
How to cite
DISEIL: Demonstration Distillation for Sample-Efficient Imitation Learning — arXiv:2609.08123 — arXiv (en), https://arxiv.org/abs/2609.08123 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f485d183f0a2 (Metraipolis archive dc_f485d183f0a2, sha256 759eb4a5f0eb1b6635bef980457357b1f4d71da0c4ac0f30ada7744d0b44260c)
BibTeX
@misc{dcf485d183f0a2,
title = {DISEIL: Demonstration Distillation for Sample-Efficient Imitation Learning — arXiv:2609.08123},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.08123},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f485d183f0a2; sha256 759eb4a5f0eb1b6635bef980457357b1f4d71da0c4ac0f30ada7744d0b44260c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08123. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A robot that can be taught a new task from a handful of demonstrations has to work out for itself what it still cannot do, and then ask for exactly that. Interactive imitation learning takes a step in that direction by letting a policy practice on its own and calling an expert when it goes wrong. Existing methods decide when to interrupt the learner. A further 2 decisions are left to whichever episode happened to trigger the interruption: which failure to correct, and where the demonstration should start. This paper is a first attempt at making both of them deliberately. DISEIL (Demonstration dIstillation for Sample-Efficient Imitation Learning) marks each failed episode at the step where the policy first becomes unreliable, represents that moment with a geometric descriptor, and groups the failures into recurring failure modes. A vision-language model and a language model read the selected mode and write a request for the next demonstration, and a store of task constraints checks that the request can be carried out before any expert time is spent. No model produces a robot action. Across 5 simulated tasks under state and image observations, changing only what the expert is asked for gives the highest mean held-out success rate in all 10 settings, with a tie in 1, and the margin is widest at the smallest budget we tested. The scope is narrow: a single round of practice at a time, in simulation, with experts that are mostly scripted. The longer-term aim is a learner that also tracks what its demonstration set already covers, and that asks a human teacher for the missing behavior in proportion to the effort each request costs them. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.08123 — 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 2026-10-16T13:39:43.256Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).