Research/Blog
Guide to Deep Learning term for the novice
- November 6, 2017
- Posted by: CellStrat Editor
- Category: Artificial Intelligence Bots Machine Learning Robotics
Deep Learning has gained significant momentum in recent months and people often confuse and use Deep Learning (DL), Machine Learning (ML) and Artificial Intelligence (AI) interchangeably. However, ML is the means to achieve AI and when we go into deeper intricacies of problem solving like face recognition etc., It’s called Deep Learning (DL). In DL too, there are quite a few terms that are used commonly but not understood easily by novice. So, here is my attempt to provide meaning to some of these in plain English so these could be understood relatively easily:
Perceptron
Human brains have trillions of biological neurons that continuously process and transmit information. Similarly, in DL, there are multiple layers of numerous perceptrons. Thus, a perceptron can be considered a super simplified version of a biological neuron.
Artificial Neural Networks
Artificial Neural Networks (ANN) are models influenced by biological neural networks such as the central nervous systems of living beings and most distinctly, the brain.
ANN’s are processing devices, such as algorithms or physical hardware, and are loosely modelled on the cerebral cortex of mammals, though on a considerably smaller scale. So, we can call them a simplified computation model of the human brain.
Back-propagation
A neural network learns by training, using an algorithm called back-propagation. To train a neural network it is first given an input which produces an output. The first step is to teach the neural network what the correct, or ideal, output should have been for that input. The ANN can then take this ideal output and begin adapting the weights to yield an enhanced, more precise output (based on how much they contributed to the overall prediction) the next time it receives a similar input.
This process is repeated multiple times until the margin of error between the input and the ideal output is considered acceptable.
Convolutional Neural Networks
A convolutional neural network (CNN) can be considered as a neural network that utilizes numerous identical replicas of the same neuron. Its benefit is that it enables a network to learn a neuron once and use it in numerous places, simplifying the model learning process in-turn reducing error. This has made CNNs particularly useful in object recognition and image tagging.
CNNs learn more and more abstract representations of the input with each convolution. In the case of object recognition, a CNN might start with raw pixel data, then learn highly discriminative features such as edges, followed by basic shapes, complex shapes, patterns and textures.
Recurrent Neural Networks
Recurrent Neural Networks (RNN) make use of sequential information. Unlike traditional neural networks, where it is assumed that all inputs and outputs are independent of one another, RNNs are reliant on preceding computations and what has previously been calculated. RNNs can be conceptualized as a neural network unrolled over time.
Recursive Neural Networks
A Recursive Neural Network is a generalization of a Recurrent Neural Network and is generated by applying a fixed and consistent set of weights repetitively, or recursively, over the structure. Recursive Neural Networks take the form of a tree, while Recurrent is a chain. Recursive Neural Nets have been utilized in Natural Language Processing for tasks such as Sentiment Analysis.
Supervised Neural Network
Supervised neural networks are neural networks that have been trained with previous inputs and desired outputs. Mostly businesses these days lack the availability of data and so it becomes difficult to solve their problem with AI. Best use cases are produced when there is huge volumes of input and desired output data is available so neural nets can be trained on this and achieve desired productivity and speed…
Unsupervised Neural Network
This involves providing a program or machine with an unlabelled data set that it has not been previously trained for, with the goal of automatically discovering patterns and trends through clustering.
(Ref: Inspired from a post on Data Science Central)