pretraining-is-transfer-learning-at-scale

IN derived (depth 2)

Created 2026-06-21T10:13:05+00:00 · Reviewed 2026-06-21T15:37:01+00:00

Modern self-supervised pretraining is transfer learning at industrial scale — the formal transfer learning framework (source domain D_S → target domain D_T) exactly describes the pretrain-then-finetune pipeline, unifying a 50-year-old theoretical concept with the dominant modern training methodology.

Justifications

SL — depth-2 — the formal transfer learning definition (base) maps directly onto the dominant pretraining pipeline (depth-1), revealing them as the same paradigm at different scales

Antecedents (all must be IN):

  • IN transfer-learning-formal-definition-domain-task — Transfer learning formally involves source domain D_S with task T_S and target domain D_T with task T_T, where either the domains or tasks must differ; if both are identical it is not transfer learning.
  • IN pretraining-finetune-dominant-but-fragile — The pretrain-then-finetune paradigm is dominant for modern deep learning — underpinning both BERT and GPT — but is empirically fragile, as pretraining can actually hurt performance when strong data augmentation or self-training alternatives are available.
  • IN self-supervised-learning-dominant-pretraining-paradigm — Self-supervised learning is the dominant pre-training paradigm for modern deep learning, as opposed to supervised or unsupervised learning.

Dependents

These beliefs depend on this one: