genetic algorithms
Genetic algorithms are inspired by the theory of evolution of species in nature. An artificial population of solutions to a problem is generated and evolved by selecting and combining the most promising solutions.
[close the glossary]
genetic programming
A branch of evolutionary computation. The main characteristic is that the population is composed of programs written in a given programming language. The main goal is to automatically generate programs for a specific task
[close the glossary]
induction
A logic process that generalizes enumerative concepts, starting from classification and comparison of single examples.
[close the glossary]
inference
Logic conclusion of a process that start from a knowledge base as a premise. Inference is an expert system feature.
[close the glossary]
lisp (list processing)
A programming language based on the concept of function. All data structures managed by this language are lists.
[close the glossary]
logic
Logic is a science that provides the basic tools for checking the correctness of reasoning. It provides formal tools for: analyzing inferences in terms of operations on symbolic expressions; deriving logical consequences from given premises; studying if a proposition is true or false depending on other propositions; checking the consistency and correctness of a theory.
[close the glossary]
logic programming
A programming paradigm that substantially differs from the traditional programming languages like C, Pascal, Fortran etc. In traditional languages, a program is a sequence of instructions that the computer should execute. In logic programming, the program is a set of logical formulae that describe the problem. The program does not contain the knowledge on how the problem should be solved.
[close the glossary]
(machine) learning
Machine learning is one of the fundamental areas of Artificial Intelligence. It concerns systems that learn new knowledge on the basis of observed examples.
[close the glossary]
min-max
Game theory algorithm for games with two players. One player should
maximize his score, while the second should minimize it. The min-max
works in cases of complete knowledge. Chess is an example of game that can be tackled with the min-max algorithm since the state is completely known by both players. Bridge is not since one player does not know the cards of every other player.
[close the glossary]
neural-network
A system composed of artificial neurons and is particularly suited for learning tasks. For example, it is extremely successful in classification. Most of the learning algorithms for neural networks adapt connections between neurons.
[close the glossary]
neuron
Cell that constitutes the nervous system; in the human brain there are approximately 100 billions neurons. Each neuron is integrated in nervous circuits. We have electric signals within each cell. Electric signals are converted into chemical messages in the communication between different cells.
[close the glossary]
neuroscience
A category of science concerning the nervous system.
[close
the glossary]
perceptron
Model of neuron behavior. Perceptrons are used as a fundamental component of many neural networks.
[close the glossary]
prolog (programming in logic)
The most famous and widely used logic programming language.
[close the glossary]
- S -
swarm intelligence
Artificial systems inspired by the behavior of social insects. These systems are composed of many "artificial insects" also called "agents", or they simulate their activity on a computer. These agents are simple, but their coordination leads to complex emerging behavior. They are able, for example, to retrieve objects, to find food and so on.
[close the glossary]
tree
A tree-shaped data structure where the nodes describe the possible states
of the system and the arches describe the possible operators that
transform these states. The tree contains all possible solutions to the
problem and is generated dynamically by the AI systems during the
calculation of the solution.
[close the glossary]
Turing test
The Turing Test introduced by Alan Turing in the paper "Computing machinery and intelligence", 1950, is a way to determine if a computer is able to think.
[close the glossary]
- W -
