Instance-based learning
Overview
Instance-based learning (also known as memory-based or lazy learning) involves learning from training examples and generalizing to new instances based on similarity measures12.
Pros
- Flexibility: Can adapt to new data without retraining.
- Simplicity: Easy to implement and understand.
Cons
- Time Complexity: High time complexity, especially with large datasets.
- Storage Requirements: Requires storing all training data.
Examples
- K-Nearest Neighbors (KNN)
- Locally Weighted Learning (LWL)
- Case-Based Reasoning (CBR)