seniorSupervised Learning
What is online learning in supervised machine learning?
Updated May 17, 2026
Short answer
Online learning updates models incrementally as new data arrives.
Deep explanation
Unlike batch learning, online learning processes data sequentially, updating the model continuously. It is useful for streaming data and dynamic environments. Algorithms like SGDClassifier support partial_fit, enabling incremental updates without retraining from scratch.
Real-world example
Real-time fraud detection systems updating models as transactions occur.
Common mistakes
- Assuming batch-trained models can adapt without retraining.
Follow-up questions
- What is concept drift in online learning?
- Which algorithms support online learning?