Metraipolis › archive › dc_30c128ba86fc

Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems — arXiv:2609.39800

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39800
License (as archived)
CC0-1.0
Archived
2026-10-07T13:46:37.842Z
SHA-256 of the archived text
830d2a05bfd25878e81fc91b39e6227e0f11a3816a76b5a0191e15813720f4d0
Stable URL of this page
https://metraipolis.com/documento/dc_30c128ba86fc

How to cite

Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems — arXiv:2609.39800 — arXiv (en), https://arxiv.org/abs/2609.39800 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_30c128ba86fc (Metraipolis archive dc_30c128ba86fc, sha256 830d2a05bfd25878e81fc91b39e6227e0f11a3816a76b5a0191e15813720f4d0)
BibTeX
@misc{dc30c128ba86fc,
  title = {Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems — arXiv:2609.39800},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39800},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_30c128ba86fc; sha256 830d2a05bfd25878e81fc91b39e6227e0f11a3816a76b5a0191e15813720f4d0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39800. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We analyse allocation, admission and post-write retention in finite-horizon linear-Gaussian noisy recurrent memories. At every horizon, the directional Fisher memory $M_n$ satisfies $\operatorname{tr}M_n=N$: non-normality redistributes information but cannot raise its spherical average, while normal carriers satisfy $M_n=I$. For bi-power-bounded carriers, we derive uniform $1/n$ lag bounds, identify the limit of $M_n$ with the inverse of the classical Cesàro asymptotic limit of $W^\top$, and give finite-horizon error bounds. A time-varying coupling defines an end-to-end store operator. The writer-optimal direction need not be store-optimal. After writing ends, an invertible hold preserves the full stored Fisher matrix. Additive contamination bounded by $\alpha$ times the closure covariance retains at least $1/(1+\alpha)$ of that matrix; a covariance-aware decoder attains the corresponding accuracy. With recurrent carriers held fixed, training input masks and linear readouts approached the task-specific optimum in 160 runs, with median normalized Rayleigh efficiency above $0.998$. Binary accuracy matched the Gaussian prediction to mean absolute error below $0.002$ over more than four orders of magnitude in $J$. In a separate pre-specified study of 320 runs, trained masks followed the designated input-time objective in both carrier types, in 16 of 16 draws. These studies used development-seen carriers and are pre-specified validations, not blind holdouts. The same fixed design reproduced the objective-specific result in 16 of 16 draws on carriers unused before run commitment. Exact isolation preserved information, while a decoder fixed at its training horizon fell to chance; inverse-adjoint transport restored its sampled decisions to numerical precision.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.39800 — 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 2026-10-07T13:18:10.456Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).