# What is Transfer Learning? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Model Training  
> **Canonical Citation:** [https://spidits.com/ai-glossary/transfer-learning](https://spidits.com/ai-glossary/transfer-learning)

## Definition
Transfer Learning is a machine learning technique where a model developed for one task is reused as the starting point for a model on a second, related task, significantly reducing the amount of labeled data and compute needed.

## Architectural Mechanics & Deep Dive
Transfer learning is a machine learning technique where a model trained on one task is repurposed as the starting point for a model on a second, related task. In AI, this is implemented by pre-training a foundation model on general text, and then fine-tuning it on specific domains, saving massive compute resources.

## Industry Applications & Real-World Use Cases
Fine-tuning foundational models (like BERT or GPT) on domain-specific datasets, and applying pre-trained ImageNet models to medical image analysis.

## Frequently Asked Questions

### What is the primary advantage of Transfer Learning?
It allows developers to train highly accurate models on very small datasets because the model starts with pre-learned features (like edge detection in images or grammar/world knowledge in text).

### What is negative transfer in Transfer Learning?
Negative transfer occurs when the knowledge learned from the source task impairs performance on the target task, usually because the two tasks are too dissimilar.

## Related Terminology
- [fine-tuning](https://spidits.com/ai-glossary/fine-tuning)
- [pre-training](https://spidits.com/ai-glossary/pre-training)
- [foundation-model](https://spidits.com/ai-glossary/foundation-model)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*