Judged: two independent readers; copy check clear.

pruning

pruning Pruning is the selective removal of parts of a model, such as channels or parameters, to make it smaller and more efficient while trying to preserve accuracy.

Pruning is the selective removal of parts of a model, such as channels or parameters, to make it smaller and more efficient while trying to preserve accuracy. In this source, adaptive pruning adjusts how much is removed based on feedback so communication and computation demands are reduced.

Sources

Drafting source: Guo Jie; Liu Renjing; Xing Jinsheng. Threshold-adaptive pruning with multi-key homomorphic encryption for communication-efficient secure federated learning. 2026; DOI 10.1371/journal.pone.0349432; article XML SHA-256 932f2078fbe149421b12af8c6f75c95da9895d898e946d3ac7f3b770479c24bc. “Threshold-adaptive pruning with multi-key homomorphic encryption for communication-efficient secure federated learning” (10.1371/journal.pone.0349432).
CC-BY-4.0 · Definition uses new wording; original source retains its stated licence. · Source snapshot SHA-256: 932f2078fbe149421b12af8c6f75c95da9895d898e946d3ac7f3b770479c24bc

These citations identify the drafting sources. 2/2 independent reader passes against the pinned source packet. The drafter does not count as a reader. Source institutions do not endorse this dictionary.

Cite this entry

PsychTerms (psychterms.org), CC BY 4.0, version 1. “pruning.” Development preview.

Version 1 · SHA-256: f75b8280783ba50f16a1696f7aa122d7f403cff71e354fe309331ca66d59ddca

Edit this entry on GitHub

pruning

See also

Sources

Cite this entry