| Evolution |
The gradual development of something, especially from a simple to a more complex form. |
| Computing |
The use or operation of computers. |
| Machine Learning |
A type of artificial intelligence that allows software applications to learn from the data and become more accurate in predicting outcomes without human intervention. |
| Genetic Algorithm |
A heuristic search and optimization technique used in artificial intelligence and computing. GSAs are commonly used to generate high-quality solutions to optimization and search problems by relying on bio-inspired operators such as mutation, recombination, and selection. |
| Neural Networks |
A computer system modeled on the human brain and nervous system, consisting of artificial neurons and synapses that process and transmit information. |
| Swarm Intelligence |
A collective behaviour that emerges from the interactions between individual agents that are not necessarily fully aware of the larger context within which they are operating. |
| Evolutionary Agent |
An agent whose internal parameters are modified through the mechanisms of natural selection, mutation, and inheritance with the goal of improving its performance on some metric. Examples of such agents include artificial life, swarm robots, autonomous cars, and other autonomous systems in general. |
| Fitness Function |
A function that is used to evaluate the performance of a given agent or model in a given task. The fitness function provides a quantifiable measure of how well the agent/model is doing and serves as the basis for the genetic operators (such as selection, mutation, and crossover) used in evolutionary algorithms. |
| Crossover |
A genetic operator used to evolve new candidate solutions from existing ones by exchanging (~crossing) randomly selected subparts of the solutions. Crossover is a key feature of genetic algorithms that allows for the propagation of good solutions across generations. |
| Mutation |
A genetic operator used to introduce random perturbations to the current candidate solution. Mutation is essential to prevent the premature convergence of the population to local optima and to allow for the exploration of new regions of the search space. |
| Selection |
A genetic operator used to select the best-fit candidates from the current population, based on their fitness values. Selection is a key step in the reproduction process of evolutionary algorithms, and it helps ensure that good solutions are propagated across generations. |
| Population |
In evolutionary computing, the population is a set of candidate solutions that are iteratively refined over multiple generations. A larger population size increases the diversity of the search process and is generally beneficial to the performance of the algorithm. |
| Decision Trees |
A decision tree is a type of supervised learning algorithm that is mostly used for classification problems. Given an input data point, a decision tree repeatedly splits the data along the dimensions that best separate the classes until it arrives at a classification decision. Thus, the decision tree is a non-parametric model. |
| Classifier |
A function that takes an input (e.g. an image) and assigns it to one of several predefined categories or classes (e.g. cat, dog, bird). In machine learning, classifiers are usually learned from labeled training data and are used to automatically label new unseen data. |
| Reinforcement Learning |
Reinforcement learning is a type of machine learning in which an agent learns to behave in an environment, by performing actions and receiving rewards or punishments. Reinforcement learning is used in scenarios where feedback from the environment is sparse, delayed, and/or costly. |
| Stochastic |
In the context of evolutionary computing, stochastic refers to probabilistic or random processes that introduce diversity and variability into the search process. Stochasticity is essential to explore new regions of the search space and avoid premature convergence to local optima. |
| Optimization |
The process of finding the best (e.g. maximum or minimum) value of some objective function, subject to some constraints. Optimization is a key area of research in evolutionary computing and has numerous applications in engineering, finance, logistics, and other fields. |
| Survival of the Fittest |
A concept in evolutionary biology that refers to the idea that organisms that are better adapted to their environment are more likely to survive and reproduce than those that are less well adapted. The concept is also used in evolutionary computing as the basis for the selection operator. |
| Multi-objective Optimization |
A type of optimization problem in which multiple conflicting objectives must be simultaneously optimized. Multi-objective optimization is an important area of research in evolutionary computing and has applications in engineering, design, and decision-making. |
