How to interpret Self Organizing Maps? A List might be better…

Self Organizing Map (SOM) is a generic method that has been applied in many applications. I spent almost 40-60 percent of my time over last three years on playing with this computational machine. Today while I was writing my PhD thesis I was trying to find a unique story line for SOM that fits well with the concept of “Pre-specific Modeling”, which is the title of my thesis. I have several technical writings to explain it in detail, but it is not convincing. As a result, I decided that may be it is better not to find a fixed way to explain SOM and maybe it should remain “pre-specific” and in this case “observer dependent”. Here, I collected a set of key words about SOM.

In principle, I can write about all of these indexes or refer to those people and papers who are focused on each of them, but now I prefer leave them as a list. Just to mention that now I am more interested in the third part of indexes which goes into analogies between SOM and other concepts.

  • Functions
    • Vector Quantization
    • Dimensionality Reduction
    • Manifold Learning
    • Topology Preservation
    • Representation Learning
    • Computing with Contextual Numbers
    • Multi-dimensional Sorting
    • Nonlinear Eigen form
    • Unsupervised Learning
    • Space Transformation
    • Visualization of High Dimensional Data
    • Data Reduction and Abstraction
    • Clustering
    • Prototype Generation and Emergence
    • Classification
    • Prediction and Function Approximation and Time Series
    • Structure Learning
    • Multi-criteria Optimization
    • Reinforcement Learning
    • Nonparametric Joint Probability Distribution and Resampling
    • Topological Data Analysis
  • Types
    • Fixed Topology
    • One-dimensional SOM
    • Two-dimensional SOM
    • Spherical SOM
    • Parametric SOM
    • Growing SOM
    • Hierarchical SOM
    • Neural Gas
    • Generative Topographic Map
    • Vector Quantization
    • Mixture of Gaussians
    • SOM as a Two layer Neural Network
    • SOM as Radial Basis Function
    • Online Learning
    • Batch Learning
    • Recurrent SOM
    • Recursive SOM
    • Relational SOM
    • Median SOM
    • Semantic SOM
    • WEBSOM
    • Vectorial representation or Non-vectorial SOM
    • Similarity Measures
    • U-Matrix
    • P-Matrix
  • Analogy
    • Associative Memory
    • Brain Cortex Analogy
    • Wave Particle Dualism
    • Mind-Matter Dualism
    • Cartesian Dualism
    • Totalitarian Geometry, Democratic Polynomials and Social Computing of SOMs
    • Geometry: Global Structures, No Local Adaptation, Polynomials: Global structure, Local Adaptation, SOM: Local Structure Local Adaptation
    • Emergent SOM: Engendering Probabilistic Concepts by Increasing the Population of Samples vs. Data Reduction Toward Ideal Representatives

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s