Metraipolis › archive › dc_112a3748cd0b

Should Pseudo-Labels Share Mini-Batches with Labeled Examples? Semi-Supervised Continual Active Learning with Limited Training Cost — arXiv:2610.35345

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.35345
License (as archived)
CC0-1.0
Archived
2026-10-27T08:02:40.588Z
SHA-256 of the archived text
d18069833331313f2f85cfccd2eca26960fdea96175fef696b7fb2bd2fd01191
Stable URL of this page
https://metraipolis.com/documento/dc_112a3748cd0b

How to cite

Should Pseudo-Labels Share Mini-Batches with Labeled Examples? Semi-Supervised Continual Active Learning with Limited Training Cost — arXiv:2610.35345 — arXiv (en), https://arxiv.org/abs/2610.35345 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_112a3748cd0b (Metraipolis archive dc_112a3748cd0b, sha256 d18069833331313f2f85cfccd2eca26960fdea96175fef696b7fb2bd2fd01191)
BibTeX
@misc{dc112a3748cd0b,
  title = {Should Pseudo-Labels Share Mini-Batches with Labeled Examples? Semi-Supervised Continual Active Learning with Limited Training Cost — arXiv:2610.35345},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.35345},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_112a3748cd0b; sha256 d18069833331313f2f85cfccd2eca26960fdea96175fef696b7fb2bd2fd01191},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.35345. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Active learning (AL) reduces annotation by querying informative examples. Continual active learning (CAL) also cuts the training cost, the total number of gradient steps, by training one model across AL rounds with a fixed number of mini-batch steps per round. Pseudo-labels add supervision without annotation, but they increase training cost. In CAL, where every round trains on a fixed number of mini-batches, whether and how pseudo-labeled examples should share these mini-batches with labeled examples is therefore open. We provide the first formalization of this problem as two decisions after each query, namely (i) which unlabeled examples should enter training with a pseudo-label and (ii) what share of each mini-batch they should take. Our method, semi-supervised continual active learning (SSCAL), works with any acquisition function. It admits pseudo-labels based on distances from unlabeled examples to labeled neighbors, caps their share, and reweights the loss on labeled examples to keep their expected gradient contribution unchanged. On frozen embeddings, its pseudo-labeler updates incrementally without extra model forward passes. Across two audio datasets, two frozen encoders, six acquisition functions and ten seeds, SSCAL ranks first on average among four protocols with equal training steps, answering the question affirmatively.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Sound (cs.SD); Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2610.35345 — 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 2026-10-27T07:55:30.985Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).