Metraipolis › archive › dc_917eeee2798c

Uniform Herding: Exemplar Replay with Representation Refresh — arXiv:2608.13061

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.13061
License (as archived): CC0-1.0
Archived: 2026-10-19T04:17:29.828Z
SHA-256 of the archived text: 760dfccb0e6756aa737e9431cfed28a03a052efbdaafa3b661209cec55eb0d1a
Stable URL of this page: https://metraipolis.com/documento/dc_917eeee2798c

How to cite

Uniform Herding: Exemplar Replay with Representation Refresh — arXiv:2608.13061 — arXiv (en), https://arxiv.org/abs/2608.13061 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_917eeee2798c (Metraipolis archive dc_917eeee2798c, sha256 760dfccb0e6756aa737e9431cfed28a03a052efbdaafa3b661209cec55eb0d1a)

BibTeX

@misc{dc917eeee2798c,
  title = {Uniform Herding: Exemplar Replay with Representation Refresh — arXiv:2608.13061},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13061},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_917eeee2798c; sha256 760dfccb0e6756aa737e9431cfed28a03a052efbdaafa3b661209cec55eb0d1a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13061. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As the feature representation changes, replay must preserve the earlier classes. However, only a bounded active exemplar set can be replayed. We propose Uniform Herding, which allocates the current active set across observed classes and uses a bounded candidate pool to refresh their chosen exemplars in the current representation. On CIFAR-100 with ten class-incremental tasks, a ResNet-18 backbone, active budget $M=2{,}000$, retrieval budget $b=64$, and three seeds, Uniform Herding obtains $44.00\pm0.51\%$ final average accuracy and $17.22\pm0.43\%$ forgetting, compared with $42.33\pm1.20\%$ and $24.87\pm1.11\%$ for iCaRL. Within the Uniform Herding protocol, final accuracy decreased when NME or herding was replaced with the tested alternatives, while forgetting increased when distillation was removed. Changing the retrieval budget has a smaller effect across the tested range than changing the active budget. The comparison with iCaRL is end-to-end. It does not isolate the effect of refresh from the other protocol differences. These results are limited to the tested protocol.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.13061 — 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 2026-10-19T04:13:06.572Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).