Unsupervised Learning

Self-organizing map

  • unsupervised machine learning technique used to proudce a low-dimensional (2D) representation of a higher dimensional data set while preserving the topological structure of the data.
  • SOM is a thpe of artificial neural network but is trained using competitive learning rather than error-correction learning (e.g., backprop wih gradient descent)
  • competitive learning: form of unsupervised learning in ANN, in which nodes compete for the right to respond to a subset of the input data. a variant of hebbian learning, competitive learning works by increasing the specialization of each node in the network. it is well suited to finding clusters within data.

UMAP

  • general purpose manifold learning and dimension reduction algorithm (compatible wit scikit-learn)