Deep learning or Deep Machine Learning is a set of algorithms in machine learning that attempts to model high-level abstractions using data architectures. We talked about Machine Learning and Artificial Intelligence through previously published articles. Before reading about Deep Learning, it is important to understand the basics from the linked articles.
What is Deep Learning or Deep Machine Learning?
Deep learning is part of a larger machine learning methods based on data representations learn together. An observation (for example, an image) can be represented in many forms (e.g., a vector of pixels ), but some representations are easier to learn tasks of interest (e.g., is this a human face?) based to examples, and research in this area attempts to define what representations are better and learn how to create models for these performances.
Several deep learning architectures, as deep neural networks, deep convolutional neural networks, deep belief networks, have been applied to fields such as computer vision, automatic speech recognition and recognition of audio and music and have been shown to produce leading results on various tasks.
There is no single definition of deep learning. In general, this is a class of algorithms for machine learning. From this common point, various publications focus on a different set of characteristics, for example; a cascade of layers of nonlinear processing units is used to extract and process characteristics. Each layer uses the output of the previous layer as input. The algorithms can use supervised or unsupervised learning, and applications include pattern recognition and statistical classification. They are based on learning (unsupervised) multi-level features or data representations. The higher level features are derived from the characteristics of lower level to form a hierarchical representation.
All of these ways of defining deep learning have in common are the multiple layers of non-linear processing and supervised or unsupervised representations learning characteristics in each layer. The layers form a hierarchy of features from a lower level to a higher abstraction.
Deep learning algorithms contrast with learning algorithms shallow by the number of transformations applied to the signal as it propagates from the input layer to the output layer. Each of these transformations includes parameters that can be trained as weights and thresholds. There is a de facto standard for the number of transformations (or layers) which turns a deep algorithm, but most researchers in the field think that deep learning involves more than two intermediate transformations.
What is Deep Learning in Human Brain
The Deep Learning we were talking about is computational deep learning. This Computational Deep Learning is closely related to one theory of neocortical development. This seem analogous to a view of the brain’s neocortex as a hierarchy of filters.
The importance of deep learning has newly regained attention with the advent of CRISPR/Cas9. CRISPR/Cas9. RNA-guided human genome engineering via Cas9 can be done among the various possibilities of CRISPR/Cas9. Among many of the dangerous works which can be done by some companies and their paid “research workers” is manipulation of a newborn’s way of natural learning.