Instance-based Learning

Instance-based Learning

Instance-based learning is a type of machine learning paradigm that operates by comparing new problem instances with instances seen in training. It is also known as memory-based learning or lazy learning, as it delays the generalization process until prediction time. This approach is fundamentally different from other learning methods, which build a general model during training to apply to new instances.


Instance-based learning algorithms store training examples and delay the induction or generalization process until a new instance must be classified or a prediction made. These algorithms use a similarity measure to identify the instances in the training data that are closest to the new instance, and base their prediction on the values or class labels of these nearest neighbors.

Key Concepts

Lazy Learning

Lazy learning is a learning method in which generalization beyond the training data is delayed until a query is made to the system. This contrasts with eager learning, where the system tries to generalize the training data before receiving queries.

K-Nearest Neighbors (K-NN)

K-NN is a popular instance-based learning algorithm. Given a new instance, K-NN searches the training set for the K instances that are closest to the new instance (according to a distance measure), and assigns the most common class among these K instances to the new instance.

Distance Measures

Distance measures are used to identify the instances in the training set that are most similar to a new instance. Common distance measures include Euclidean distance, Manhattan distance, and Minkowski distance.

Feature Weighting

In many instance-based learning algorithms, some features are more important than others for determining the class of a new instance. Feature weighting methods assign weights to features according to their importance.

Advantages and Disadvantages

Instance-based learning has several advantages. It makes no assumptions about the underlying data distribution, making it suitable for complex problems. It can also adapt quickly to changes, as it does not require retraining to incorporate new data.

However, instance-based learning also has some disadvantages. It requires a large amount of memory to store the training instances. It can also be computationally expensive, as it needs to compute the distance to all training instances for each new instance. Furthermore, it is sensitive to the choice of distance measure and the value of K in K-NN.


Instance-based learning is used in various fields, including computer vision, pattern recognition, and recommendation systems. It is particularly useful in situations where the data may change rapidly, or where the relationships between features are complex and hard to model with traditional methods.

Further Reading

For more in-depth information on instance-based learning, consider reading the following resources:

  • Aha, D.W., Kibler, D., & Albert, M.K. (1991). Instance-based learning algorithms. Machine Learning, 6, 37-66.
  • Cover, T., & Hart, P. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21-27.
  • Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning. Springer.

This glossary entry was last updated on August 14, 2023.