Unsupervised learning is a type of machine learning that looks for previously undetected patterns in a data set with no pre-existing labels and with a minimum of human supervision. In contrast to supervised learning that usually makes use of human-labeled data, unsupervised learning, also known as self-organization allows for modeling of probability densities over inputs.

Here are the main aspects of unsupervised learning:

1. Clustering: This is about grouping a set of objects in such a way that objects in the same group are more similar to each other than to those in other groups. Examples include K-means, hierarchical clustering, and Gaussian mixture models.
2. Association: This method is used to find rules that will describe large portions of data, such as the Apriori algorithm or Market Basket Analysis.
3. Dimensionality Reduction: Methods like Principal Component Analysis (PCA) or t-Distributed Stochastic Neighbor Embedding (t-SNE) are used to reduce the number of random variables to consider, with a particular focus on visualizing high-dimensional datasets.
4. Anomaly Detection: Unsupervised learning can be used to detect abnormal patterns that do not conform to expected behavior, such as fraud detection or fault detection.
5. Neural Networks: Some neural networks can be trained using unsupervised learning techniques, like Autoencoders.

Unsupervised learning can be incredibly powerful when dealing with large amounts of unlabelled data. It can find complex patterns and can be used to prepare data for supervised learning, identify outliers, or reduce dimensionality. However, it can be more unpredictable and challenging to understand compared to supervised learning, where the learning process can be more easily guided and interpreted.