[Ref: http://en.wikipedia.org/wiki/Deep_learning]


a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using model architectures, with complex structures or otherwise, composed of multiple non-linear transformations.

  • use a cascade of many layers of nonlinear processing units for feature extraction and transformation
  • are based on the (unsupervised) learning of multiple levels of features or representations of the data, the levels form a hierarchy of concepts.

Deep learning algorithms are based on distributed representations.Deep learning adds the assumption that these factors are organized into multiple levels, corresponding to different levels of abstraction or composition. Varying numbers of layers and layer sizes can be used to provide different amounts of abstraction.

rather than focusing on feature engineering which is often labor-intensive and varies from one task to another, deep learning methods is focused on end-to-end learning based on raw features.

Deep learning architectures:

  • deep neural networks.

DNNs are typically designed as feedforward networks. A DNN can be discriminatively trained with the standard backpropagation algorithm.The weight updates can be done via stochastic gradient descent using the following equation:

$\Delta w_{ij}(t + 1) = \Delta w_{ij}(t) + \eta\frac{\partial C}{\partial w_{ij}}$

  • convolutional deep neural networks

A CNN is composed of one or more convolutional layers with fully connected layers (matching those in typical artificial neural networks) on top. CNNs are easier to train than other regular, deep, feed-forward neural networks and have many fewer parameters to estimate

  • deep belief networks

A deep belief network (DBN) is a probabilistic, generative model made up of multiple layers of hidden units. A DBN can be used for generatively pre-training a DNN by using the learned weights as the initial weights.

  • Deep Boltzmann Machines