What is Autoencoder? An autoencoder is an unsupervised deep learning algorithm used for: Feature detection. When an autoencoder encodes data, the hidden layer represents important features, whi...
Boltzmann Machine
What is Boltzmann Machine? A Boltzmann machine is an unsupervised deep learning model used for recommendation systems. Here is a simple structure of a Boltzmann machine: There are four signifi...
Self Organizing Map(SOM)
What is SOM? SOM is an unsupervised deep learning technique used for dimensionality reduction. Here is an example of SOM in use: How does SOM learn? Here is a simple example of SOM stru...
Activation Functions
What are Activation Functions? Activation functions are used in 3rd step in the image below. Types of Activation Functions There are 5 types of activation functions: Threshold function The...
Recurrent Neural Network(RNN)
What is RNN? RNN is a type of supervised deep learning used for time series analysis. The basic structure of RNN looks like: The hidden layers (the blue circles) not only give output but also f...
Gradient Descent
What is Gradient Descent? Gradient Descent is a method to minimize the cost $C$. The formula for the cost $C$ is: $\frac{1 \times (\hat{y}-y)^{2}}{2}$ The graph of $C$ is shown below. Let...
Convolutional Neural Network (CNN)
What is CNN? A CNN is trained on classified images and makes predictions based on them. After going through a CNN consisting of Convolution, Pooling, Flattening, and Full Connection steps, im...
Artificial Neural Networks (ANN)
What is an ANN? ANNs are computing systems inspired by the biological neural networks that constitute animal brains. ANNs are based on a collection of connected units or nodes called artificial neu...
Kernel Principal Component Analysis (Kernel PCA)
What is Kernel PCA? PCA is a linear algorithm whereas Kernel PCA is a non-linear algorithm. Kernel PCA is used for: Noise filtering Visualization Feature extraction Stock market predicti...
Linear Discriminant Analysis (LDA)
What is LDA? LDA differs from PCA because, in addition to finding the component axises with LDA we are interested in the axes that maximize the separation between multiple classes. PCA is an unsupe...