397,759 Hits in 6.3 sec

On the definition of a general learning system with user-defined operators [article]

Fernando Martínez-Plumed and Cèsar Ferri and José Hernández-Orallo and María-José Ramírez-Quintana
2013 arXiv   pre-print
Since changing operators affect how the search space needs to be explored, heuristics are learnt as a result of a decision process based on reinforcement learning where each action is defined as a choice  ...  As a result, the architecture can be seen as a 'system for writing machine learning systems' or to explore new operators where the policy reuse (as a kind of transfer learning) is allowed.  ...  As we have mentioned in Section 1, in this paper we describe the gErl system [40] a general learning system which can be configured with different (possibly user-defined) operators and where the heuristics  ... 
arXiv:1311.4235v1 fatcat:7pqwyo7vpzhmtimdumfcfowayu

Selecting Vision Operators and Fixing Their Optimal Parameters Values Using Reinforcement Learning [chapter]

Issam Qaffou, Mohamed Sadgal, Aziz Elfazziki
2012 Lecture Notes in Computer Science  
This solution uses a multi-agent architecture based on reinforcement learning to automate the process of operator selection and parameter adjustment.  ...  Selecting the appropriate operators with the optimal values for their parameters represents a big challenge for users. In this paper we present a solution for this problem.  ...  The type of the extracted features depends on the task at hand. Here we give a general definition, and in the experience we define them explicitly according to the application.  ... 
doi:10.1007/978-3-642-31254-0_12 fatcat:gook5p6zc5evlpyzax5yuy2vgq

A Framework for the Generation of Adaptive Courses Based on Semantic Metadata

Freddy Duitama, Bruno Defude, Amel Bouzeghoub, Claire Lecocq
2005 Multimedia tools and applications  
Our approach proposes the creation and management of adaptive learning systems by combining component technology, semantic metadata, and adaptation rules.  ...  A component model allows interaction among components that share consistent assumptions about what each provides and each requires of the other.  ...  Rules are defined by domain experts. The system is initiated with a set of general rules allowing to express generic rewriting strategies.  ... 
doi:10.1007/s11042-005-6541-8 fatcat:njwstqo6ibc2rinrp4mrjt63le

Transparency of Classification Systems for Clinical Decision Support [chapter]

Antoine Richard, Brice Mayag, François Talbot, Alexis Tsoukias, Yves Meinard
2020 Communications in Computer and Information Science  
In collaboration with the Civil Hospitals of Lyon, we aim to develop a "transparent" classification system for medical purposes.  ...  Therefore, in this paper, we propose a definition of "transparency" for classification systems in medical contexts.  ...  This paper was made in collaboration with employees of the Civil Hospitals of Lyon (France). Thanks to all of them. Special thanks to Pr. Moulin and Dr.  ... 
doi:10.1007/978-3-030-50153-2_8 fatcat:auot6exnxrgwrkpxdcr4r2iu54

Adaptivity vs. the idea of progression in language learning environments: Antagonism or complementarity?

Antoine Besnehard, Armand Heroguel
2012 Procedia - Social and Behavioral Sciences  
learning process, even in an adaptive system with a high level of user-generated input or of user control.  ...  With a focus on computer systems, we read in the Encyclopedia Britannica 3 that "adaptive control is the capability of the system to modify its own operation to achieve the best possible mode of operation  ... 
doi:10.1016/j.sbspro.2012.02.007 fatcat:myzsedkx2bawdjttcjjt7pdfkq

Providing Flexible Process Support to Project-Centered Learning

S. Ceri, F. Daniel, M. Matera, A. Raffio
2009 IEEE Transactions on Knowledge and Data Engineering  
We discuss how the general paradigm of flexible processes was adapted to the learning concept, implemented, and experienced by students.  ...  Tools are most effective when they embed some knowledge about the process, e.g., in terms of the typical activities required by the process, so that users can execute the activities without having to define  ...  ACKNOWLEDGMENTS This work is founded by the FP6 EU Project COOPER (IST027073). The authors are grateful to Professor Jan van  ... 
doi:10.1109/tkde.2008.134 fatcat:5lb6il2fjbgljcqoujvz24p5fi

A New Automatic Method to Adjust Parameters for Object Recognition

Issam Qaffou, Mohamed Sadgal, Aziz Elfazziki
2012 International Journal of Advanced Computer Science and Applications  
To recognize an object in an image, the user must apply a combination of operators, where each operator has a set of parameters.  ...  Due to the nature of reinforcement learning the results do not depend only on the system characteristics but also the user's favorite choices.  ...  Here we give a general definition, and in the experience we define them explicitly for a recognition task. C.  ... 
doi:10.14569/ijacsa.2012.030932 fatcat:u545takp6zbrvd6mtodpnicxiu

Bringing APL down to earth on the DECsystem-10: Standard characters and a standard editor

Fred A. Masterson
1981 Behavior Research Methods  
This question is particularly important in light of the obvious inferiority of many APL editors to the general-purpose editor of the typical host computer system.  ...  Finally, individual lines of APL code can be interpreted outside an official function definition, a feature that encourages trial-and-error learning.  ...  In addition, APL respects spaces on either side of the names of user-defined functions.  ... 
doi:10.3758/bf03202036 fatcat:dsgy2rnsyrhixii5xzepgpyfcq

Q-learning optimization in a multi-agents system for image segmentation [article]

Issam Qaffou, Mohamed Sadgal, Abdelaziz Elfazziki
2013 arXiv   pre-print
This paper proposes a solution to this problem as a multi-agent system modeled according to the Vowel approach and using the Q-learning algorithm to optimize its choice.  ...  To know which operators to apply and in which order, as well as attributing good values to their parameters is a challenge for users of computer vision.  ...  a) State definition Generally, we define a state s i representing an image in an iteration I, by a function of a set of features : The features that can be used depend on the task at hand and the specifications  ... 
arXiv:1311.6054v1 fatcat:54tyers5snc7ve6eh2v6u7ttry

Desiging robot services with ontology and learning

Chiung-Hon Leon Lee, Alan Liu
2009 2009 IEEE International Conference on Systems, Man and Cybernetics  
However, ways of controlling robots may differ depending on users. Learning from past experiences enables a robot to adapt to the user's specific needs and interacting style.  ...  This article focuses on defining robot services that can suit user's preferences. Service robots are typically designed to provide one set of services not targeted to a particular user.  ...  Fig.6 shows the CSV file being imported to the system, and Fig.7 shows the resulting ontology definition to 'Users' with seven slots and different values defined.  ... 
doi:10.1109/icsmc.2009.5346946 dblp:conf/smc/LiuL09 fatcat:na5fqjyo2jearag2cpoosregtu

Towards an Ontology of Cognitive Assistants

Torsten Maier, Jessica Menold, Christopher McComb
2019 Proceedings of the International Conference on Engineering Design  
This paper discusses the development of an ontology, its classes, and subclasses that may serve as a foundation for defining and differentiating CAs.  ...  Our novel ontological framework is the first step towards a classification system for this burgeoning field.  ...  For example, Herron defined intelligent agents as "a form of artificial intelligence that operate on user commands" (p. 139).  ... 
doi:10.1017/dsi.2019.270 fatcat:q6ihg7hxmzazdixtjmipktx6se

SPHINX: Schema integration by example

Francois Barbançon, Daniel P. Miranker
2007 Journal of Intelligent Information Systems  
The meta-model of federating view definitions enables, first, an active-learning method which removes the burden of generating the examples from the user.  ...  Second, SPHINX can identify when the learned view definition converges to a vetted, semantically accurate result. Thus, the users do not need to verify the correctness of the transformation.  ...  Section 6 deals with the active learning portion of the SPHINX system, which generates and selects data examples to submit to the user.  ... 
doi:10.1007/s10844-006-0011-2 fatcat:f3x4g66bxvbvvlb72asg3rygqu

A network and repository for online laboratory, based on ontology

H. Saliah-Hassane, R. C. Correia, J. M. Fonseca
2013 2013 IEEE Global Engineering Education Conference (EDUCON)  
With the drawing of this new standard we pretend define methods for storing and retrieving learning objects for remote laboratories.  ...  The objects defined by this standard are, for example, interfaces for devices connected to user computers over computers networks and the devices themselves.  ...  On the other hand, if we have a simple interface system the user easily learn how to work with the interface and is easier to use and doesn't require learning and adaptation time from the user.  ... 
doi:10.1109/educon.2013.6530257 dblp:conf/educon/Saliah-HassaneCF13 fatcat:nvekl4vj4rai5kl7z7bikvv6pi

Organizing Problem Solving Activities for Synchronous Collaborative Learning of Design Domains [chapter]

Crescencio Bravo, Miguel A. Redondo, Manuel Ortega, José Bravo
2003 Lecture Notes in Computer Science  
First, its positions are generalized, and next, a specification that will allow users to generate software systems that approach other design domains is defined.  ...  Scientific community is showing a growing interest in Collaborative Systems, although most of developed systems deal with specific applications.  ...  This work was partially supported by the Junta de Comunidades de Castilla -La Mancha and the Ministerio de Ciencia y Tecnología in the Projects PBI-02-026 and TIC2002-01387.  ... 
doi:10.1007/3-540-45068-8_19 fatcat:knlcyfdqkfbyppji6j5ypqizeu

A Multi-Agents Architecture to Learn Vision Operators and their Parameters [article]

Issam Qaffou, Mohammed Sadgal, Abdelaziz Elfazziki
2012 arXiv   pre-print
The diversity of operators and the multitude of their parameters constitute a big challenge for users.  ...  Through the reinforcement learning mechanism, our architecture does not consider only the system opportunities but also the user preferences.  ...  Reinforcement Learning According to the definition of S.Sutto and G.Barto [9] , reinforcement learning defines a type of interaction between an agent and its environment.  ... 
arXiv:1207.2426v1 fatcat:pbqrpavwrzewdphhsu7msnbxry
« Previous Showing results 1 — 15 out of 397,759 results