Atlas · GenAI 2026
Transfer Learning
Transfer learning
conceptPeak: 2019Training ParadigmsAI consensus: 1/3
Prerequisites
- hardDeep Learning
Transfer learning is meaningless without understanding what pretrained representations are and how fine-tuning modifies learned features
Recommended reference
Howard & Ruder (2018) 'Universal Language Model Fine-tuning for Text Classification' (ULMFiT) — paper that launched modern transfer learning in NLP
Notes from AI deep research
Anthropic Opus
ULMFiT (2018) zapoczatkowal modern NLP transfer. Cala era LLM to transfer learning at scale
Related skills
- → is subcategory of: Machine Learning(3/3)
- ← is subcategory of: LoRA / QLoRA(1/3)