Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?

Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.

Découvrez-en plus sur les applications Graph.

Graph
Search

fr|en

Se Connecter

Concept

Agent intelligent

Résumé

En intelligence artificielle, un agent intelligent (AI) est une entité autonome capable de percevoir son environnement grâce à des capteurs et aussi d'agir sur celui-ci via des effecteurs afin de réaliser des objectifs. Un agent intelligent peut également apprendre ou utiliser des connaissances pour pouvoir réaliser ses objectifs. Ils peuvent être simples ou complexes. Par exemple, un simple système réactif, comme le thermostat est considéré comme étant un agent intelligent. Les agents intelligents sont souvent décrits schématiquement comme des systèmes fonctionnels abstraits similaires aux logiciels informatiques. C'est pour cette raison que les agents intelligents sont parfois nommés agents intelligents abstraits (AIA) pour les distinguer de leurs implémentations sous forme de logiciels informatiques, systèmes biologiques ou organisations. Certaines définitions mettent en avant leur autonomie et vont donc préférer le terme agent intelligent autonome. Références

Source officielle

https://fr.wikipedia.org/wiki/Agent_intelligent

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (87)

Chargement

Afficher plus

Publications associées

Chargement

Personnes associées (9)

Personnes associées

Chargement

Unités associées (7)

Unités associées

Chargement

Concepts associés (32)

Concepts associés

Chargement

Cours associés (14)

Software agents are widely used to control physical, economic and financial processes. The course presents practical methods for implementing software agents and multi-agent systems, supported by programming exercises, and the theoretical underpinnings including computational game theory.

This course provides an overview of the theory of asset pricing and portfolio choice theory following historical developments in the field and putting emphasis on theoretical models that help our understanding of financial decision making and financial markets.

This course introduces the theory and application of modern convex optimization from an engineering perspective.

Afficher plus

Cours associés

Chargement

Séances de cours associées (30)

Séances de cours associées

Chargement

Publications associées (87)

CoPainter: Interactive AI drawing experience

Pawel Golinski

Google AI Experiments [3] is a set of simple experiments that showcase innova- tive uses of AI in UX. As part of this collection, Google shared Quick Draw, where user draws simple sketches and intelligent agent tries to guess what it depicts, and Sketch- RNN interactive demo, an experience where user starts a sketch and recurrent neural network model completes it [4]. These experiences provide fun and engaging drawing activities which have potential to be used in more educational context. This report describes CoPainter - an interactive AI drawing experience which aims to provide an educational activity based on learning by teaching paradigm - the application asks the user to draw a painting of some object in cooperation with an intel- ligent agent (which can either be a robot or an abstract entity). The activity is based on Sketch-RNN model mentioned above. Main contributions of this project are porting the Sketch-RNN web app experience to Qt, which enables integration with CoWriter code- base and makes it possible to collect more data on user activities. An additional draw in turns game mode is also introduced. The activity may possibly be extended to enable it to be used in diagnosing and treating dysgraphia, which is briefly mentioned in this re- port and is aim of further research on this project.

2019

Chargement

Multiple Heterogeneous Ant Colonies with Information Exchange

Arash Arami, Bijan Rahimzadeh Rofoee

The method of multiple heterogeneous ant colonies with information exchange (MHACIE) is presented in this paper with emphasis on the speed of finding the optimal solution and the corresponding computational complexity. The proposed method which is inspired by biology and psychology has a structure composed of several ant colonies. These colonies participate in solving problems in a concurrently manner and also exchange information with each other in communicational steps. Each ant colony is considered as an intelligent agent with behavioral traits. These behavioral traits play a key role in the solving procedure, in interrelation circumstances and in installation of relations. Faster solutions have been achieved using different employments of agents in the algorithm structure. Experimental results show the superiority of Multiple heterogeneous ant colonies algorithm in comparison to the standard ant colony system (ACS) and particle swarm optimization (PSO) algorithms on different benchmarks. A dynamic, control engineering benchmark is also provided in order to gain a more complete evaluation of the proposed algorithm.

2008

Chargement

Dynamic Resolution in the Co-Evolution of Morphology and Control

Joshua Evan Auerbach

Evolutionary robotics is a promising approach to overcoming the limitations and biases of human designers in producing control strategies for autonomous robots. However, most work in evolutionary robotics remains solely concerned with optimizing control strategies for existing morphologies. By contrast, natural evolution, the only process that has produced intelligent agents to date, may modify both the control (brain) and morphology (body) of organisms. Therefore, co-evolving morphology along with control may provide a better path towards realizing intelligent robots. This paper presents a novel method for co-evolving morphology and control using CPPN-NEAT. This method is capable of dynamically adjusting the resolution at which components of the robot are created: a large number of small sized components may be present in some body locations while a smaller number of larger sized components is present in other locations. Advantages of this capability are demonstrated on a simple task, and implications for using this methodology to create more complex robots are discussed.

MIT Press2010

Chargement

Afficher plus