Metraipolis › archive › dc_e10baade1d19

Unified dynamics of training, learning, inference, and reinforcement in neural networks — arXiv:2611.01345

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.01345
License (as archived)
CC0-1.0
Archived
2026-11-03T18:36:50.935Z
SHA-256 of the archived text
8282139168cc458efe3606445bf3878e202cc7631d7a85bc73d8252a4ab2737b
Stable URL of this page
https://metraipolis.com/documento/dc_e10baade1d19

How to cite

Unified dynamics of training, learning, inference, and reinforcement in neural networks — arXiv:2611.01345 — arXiv (en), https://arxiv.org/abs/2611.01345 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e10baade1d19 (Metraipolis archive dc_e10baade1d19, sha256 8282139168cc458efe3606445bf3878e202cc7631d7a85bc73d8252a4ab2737b)
BibTeX
@misc{dce10baade1d19,
  title = {Unified dynamics of training, learning, inference, and reinforcement in neural networks — arXiv:2611.01345},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01345},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e10baade1d19; sha256 8282139168cc458efe3606445bf3878e202cc7631d7a85bc73d8252a4ab2737b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01345. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Despite their remarkable success, neural networks remain largely black boxes, and the mechanisms underlying their behavior are still poorly understood. Nine nanoGPT experiments established training-learning-inference relations: realized updates elicit state-dependent finite-amplitude responses; learning persistently reorganizes distributed functional support, with target-specific readout-boundary crossings determining capability transitions; inference recruits learned support without changing the learned state. Factors identified in the fifth experiment yielded a direct predictor of post-update capability transitions, achieving 91.43% accuracy and 91.49% four-way macro recall across 5,088 targets. We tested these relations and prediction across five additional neural architectures. In a seventh system, concentrated correct feedback eroded previously mastered skills; rebalanced feedback restored their accuracy from 70.31% to 99.48%. Linking inference consequences to subsequent training closes these relations into unified training-learning-inference-reinforcement dynamics.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.01345 — 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 2026-11-03T18:12:54.494Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).