Computational Intelligence

6 min readJul 3, 2019

Computational Intelligence is the study of adaptive mechanisms to enable or facilitate intelligent behavior in complex and changing environments.Enormous successes have been achieved through the modeling of biological and natural intelligence, resulting in the intelligence system. These intelligent algorithms include evolutionary computation, swarm intelligence, artificial immune system, fuzzy systems. Together with logic, deductive reasoning, expert systems,case-based reasoning, and symbolic machine learning system, these intelligent algorithms form part of the field of Artificial intelligence.

Can computers be intelligent?

In the 1900s, Alan Turing gave much thought to this question. he believed that machines could be created that would mimic the process of the human brain. while success has been achieved in modeling small parts of biological neural systems, there is still no solution to the complex problem of modeling intuition, consciousness, and emotion which form intelligent parts of human intelligence. In 1950, Turing published his test of computer intelligence, referred to as the Turing test. The test consists of the person asking questions via a keyboard to both a person and computer apart from the human, the computer could be perceived as being intelligent.

Computational Intelligence Paradigms

Artificial Neural Networks(ANN)
Evolutionary Computation(EC)
Swarm Intelligence(SI)
Artificial Immune System(AIS)
Fuzzy System(FS)

Artificial Neural Network(ANN)

The ability to learn, memorize, and still generalize, prompted research in algorithmic modeling of biological neural systems-referred to as Artificial Neural Network.

Types of ANN

Single layer NNs
Multilayer feedforward NNs
Temporal NNs
Self-organizing NNs
Combined Supervised and unsupervised learning

The Artificial Neuron

An artificial neuron is a model of a biological neuron. Each AN receives signals from the environment, or other ANs, gathers these signals, and when fired, transmits a signal to all connected ANs.Input signals are inhibited or excited through negative and positive numerical weights associated with each connection to the AN.The firing of AN and the strength of the exiting signal are controlled via a function, referred to as the activation function.

Activation functions

(a) Linear Function

(b) Step Function

© Ramp Function

(d) Figmoid Function

(e)Hyperbolic Function

(f) Gaussian Function

Artificial Neuron Learning

There are three main types of learning.

Supervised Learning

The neuron is provided with a dataset consisting of input vector and target associated with each input vector. This dataset is referred to as the training set.

Unsupervised Learning

The aim of unsupervised learning is to discover features (or) patterns in the input data with no assistance from the external source. Much unsupervised learning basically performs clustering.

Reinforcement Learning

The aim of reinforcement learning is to reward the neuron for its good performance and penalize for bad performance

Evolutionary Computation(EC)

EC mimic processes from natural evolution, where the main concept is the survival of the fittest: the weak must die.In a natural evolution, survival is achieved through reproduction. Offspring, reproduced from two parents (sometimes more than two), contain genetic material of both (or all) parents — hopefully, the best characteristics of each parent. Those individuals that inherit bad characteristics are weak and lose the battle to survive.

Example: In some bird species, where one hatchling manages to get more food, gets stronger, and at the end kicks out all its siblings from the nest to die.

Genetic Algorithm(GA)

GA is possibly the first algorithmic models developed to simulate genetic systems. The main driving operators of GA are selection and recombination through the application of a crossover operator.

Crossover

Crossover is the process of exchanging genetic material between two homologous chromosomes. Crossover operators can be divided into three main categories based on arity(ie.number of parents used) of the operators.

asexual-offspring is generated from one parent
sexual-offspring is generated from two parents.
multi parent-two or more parents id used to produce two or more offspring.

Crossover can be represented in two ways

Binary string
Floating point

Mutation

The aim of the mutation is to introduce new genetic material into an existing individual. This is used to add diversity to the genetic characteristics of the population. A mutation is used in support of crossover to ensure that the full range of alleles is accessible for each gene.

Mutation operators

Uniform
Inorder
Gaussian

Genetic Programming(GA)

Genetic programming is a specialization of genetic algorithms. Similar to GAs, GP concentrates on the evolution of genotypes.the main difference between the two paradigms is representation scheme used. where GA use string and GP uses tree representation.

Crossover operators

Generating one offspring

Generating two offsprings

Swarm Intelligence(SI)

Swarm intelligence (SI) originated from the study of colonies, or swarms of social organisms.Studies of the social behavior of organisms (individuals) in swarms prompted the design of very efficient optimization and clustering algorithms.

Example: the choreography of bird flocks led to the design of the particle swarm optimization algorithm, and studies of the foraging the behavior of ants resulted in an ant colony optimization algorithms.

Particle Swarm Optimization(PSO)

PSO is a stochastic optimization approach, modeled on the social behavior of bird flocks.PSO is a population-based search procedure where the individuals, referred to as particles are grouped into a swarm. Each particle in the swarm represents a candidate solution to the optimization problem. In a PSO system, each particle is “flown” through the multidimensional search space, adjusting its position in search space according to its own experience and that of neighboring particles. A particle, therefore, makes use of the best position encountered by itself and the best position of its neighbors to position itself toward an optimum solution. The effect is that particles “fly” toward an optimum, while still searching a wide area around the current best solution. The performance of each particle is measured by a predefined fitness.

Artificial Immune System(AIS)

The natural immune system (NIS) has an amazing pattern matching ability used to distinguish between foreign cells entering the body (referred to as non-self, or antigen) and the cells belonging to the body (referred to as self).

There are four models in NIS

1. The classical view of the immune system is that the immune system distinguishes between self and non-self, using lymphocytes produced in the lymphoid organs. These lymphocytes “learn” to bind to the antigen.

2. Clonal selection theory, where an active B-Cell produces antibodies through a cloning process. The produced clones are also mutated.

3. Danger theory, where the immune system has the ability to distinguish between dangerous and non-dangerous antigen.

4.Network theory, where it is assumed that B-Cells form a network. When a B-Cell responds to an antigen, that B-Cell becomes activated and stimulates all other B-Cells to which it is connected in the network.

Fuzzy system

The traditional set theory requires elements to be either part of a set or not. Similarly, binary-valued logic requires the values of parameters to be either 0 or 1.Fuzzy sets and fuzzy logic allows what is referred to as approximate reasoning. With fuzzy sets, an element belongs to a set to a certain degree of certainty.

Applications: control systems, gear transmission, and braking systems in vehicles, controlling lifts, home appliances, controlling traffic signals, and many others.