Metraipolis › archive › dc_1653cecbdd11
Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics — arXiv:2609.09099
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.09099
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T13:58:30.063Z
- SHA-256 of the archived text
4f055aec60aec99ed76e28480d3cc6f3ccbf4aa64b698ab13478a1161b920274- Stable URL of this page
https://metraipolis.com/documento/dc_1653cecbdd11
How to cite
Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics — arXiv:2609.09099 — arXiv (en), https://arxiv.org/abs/2609.09099 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1653cecbdd11 (Metraipolis archive dc_1653cecbdd11, sha256 4f055aec60aec99ed76e28480d3cc6f3ccbf4aa64b698ab13478a1161b920274)
BibTeX
@misc{dc1653cecbdd11,
title = {Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics — arXiv:2609.09099},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.09099},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1653cecbdd11; sha256 4f055aec60aec99ed76e28480d3cc6f3ccbf4aa64b698ab13478a1161b920274},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09099. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Curriculum learning is governed by several coupled design choices---how difficulty is defined, how examples are ordered, how much exposure each level receives, and how quickly training moves across levels---making it hard to isolate what actually helps. We present Wasserstein curriculum paths, a simple transport-based framework that decouples these factors by representing curricula as trajectories of training distributions over discrete difficulty levels. Across a calibrated synthetic suite with 12 tasks and 33 difficulty axes, we use this framework to isolate the effects of ordering, matched exposure, endpoint smoothness, and pacing under fixed training budgets. We find that curriculum effects are strongly context-dependent: no single strategy dominates across tasks, difficulty axes, and budgets, and curricula mainly change where a fixed budget is spent most effectively. Within this framework, easy-to-hard ordering improves hard-level performance relative to exposure-matched static sampling, showing that the benefit is not explained by cumulative exposure alone. We further show that endpoint smoothness and pacing substantially affect where along the difficulty spectrum a curriculum is effective. Finally, we show that the same transport view naturally supports extensions to learned pacing through geometry and to structured difficulty spaces beyond one-dimensional orderings. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.09099 — 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 2026-10-16T13:39:43.090Z 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).