Metraipolis › archive › dc_c225fc319f01

FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss — arXiv:2610.39423

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.39423
License (as archived)
CC0-1.0
Archived
2026-10-31T06:13:16.340Z
SHA-256 of the archived text
bbd3a1c4fe9d2c2e4659bfb88f29cdeedd79aa78d95e4ee90946a9e2ff778c88
Stable URL of this page
https://metraipolis.com/documento/dc_c225fc319f01

How to cite

FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss — arXiv:2610.39423 — arXiv (en), https://arxiv.org/abs/2610.39423 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c225fc319f01 (Metraipolis archive dc_c225fc319f01, sha256 bbd3a1c4fe9d2c2e4659bfb88f29cdeedd79aa78d95e4ee90946a9e2ff778c88)
BibTeX
@misc{dcc225fc319f01,
  title = {FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss — arXiv:2610.39423},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.39423},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c225fc319f01; sha256 bbd3a1c4fe9d2c2e4659bfb88f29cdeedd79aa78d95e4ee90946a9e2ff778c88},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39423. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Formula-Driven Supervised Learning (FDSL) enables vision pre-training without real images, but existing methods cast pre-training as classification on discretized labels, discarding continuity in the generative process. We propose Formula-Driven Supervised Metric Learning (FDSML), which derives relational supervision from image-generation parameters beyond discrete class labels. FDSML combines FineFractalDB, our refined fractal dataset with higher intra-class variation and more stable structures, with a Parameter-Aware Triplet (PA-Triplet) Loss that assigns positives and negatives according to distances in the underlying Iterated Function System parameter space. This objective trains embeddings to respect parameter-derived relative similarity orderings. Experiments show that FDSML improves transferability to image retrieval while remaining competitive on downstream classification and improving robustness under several corruption settings. On ResNet-50, Recall@1 improves from 65.6% to 74.0% on Stanford Online Products and from 56.2% to 78.4% on Stanford Cars over FDSL pre-training. These results suggest that parameter-derived relational supervision provides a stronger synthetic pre-training signal than label-based classification for transfer beyond recognition.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE).
Pagina del lavoro: https://arxiv.org/abs/2610.39423 — 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 2026-10-31T06:07:08.718Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).