Continual Learning as Shared-Manifold Continuation Under Compatible Shift

Henry J. Kobs · 2026

Read Continual Learning as Shared-Manifold Continuation Under Compatible Shift by Henry J. Kobs (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-20036.txt.

Description

Kobs reframes continual learning as keeping old and new data on one shared latent manifold. Standard methods fight forgetting by penalizing parameter change or replaying old samples. Those tricks never say how the internal representation should evolve. Kobs builds a method, SPMA, around preserving the shared support of the data. A geometry-aware variant adds sparse replay, output distillation, and anchor-based shape keeping. On CIFAR-10 and Tiny-ImageNet shifts, it retained old tasks better than replay baselines. A synthetic atlas test showed near-perfect preservation of anchor geometry.

Catalog record

CREATORHenry J. Kobs
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21604
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.20036

Browse the full library · About the Tower of Babel