Representation learning, also known as representation learning, is a method that uses machine learning to obtain a vectorized expression of each entity or relationship so that useful information can be more easily extracted when building classifiers or other predictive variables.

In machine learning, representation learning is a technical integration of feature learning: that is, converting raw data into a form that can be exploited by machine learning. It avoids the tedious manual extraction of features and allows learning to use features while mastering the extraction method.

Representation Learning Classification

There are two main types of representation learning: supervised representation learning and unsupervised representation learning.

• Supervised representation learning: Labeled data is used as features for learning, such as neural networks, multi-layer perceptrons, and supervised dictionary learning;
• Unsupervised representation learning: unlabeled data is used as features for learning, such as unsupervised dictionary learning, independent component analysis, autoencoding, matrix decomposition, cluster analysis and their variations.

Parent/Related Words: Machine Learning