The Experts below are selected from a list of 717 Experts worldwide ranked by ideXlab platform
Ioannis Hatzilygeroudis - One of the best experts on this subject based on the ideXlab platform.
-
SETN - Improving the Integration of Neuro-Symbolic Rules with Case-Based Reasoning
Artificial Intelligence: Theories Models and Applications, 2008Co-Authors: Jim Prentzas, Ioannis Hatzilygeroudis, Othon MichailAbstract:In this paper, we present an improved approach integrating rules, neural networks and cases, compared to a previous one. The main approach integrates neurules and cases. Neurules are a kind of integrated rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. Each neurule is represented as an adaline unit. The main characteristics of neurules are that they improve the performance of Symbolic rules and, in contrast to other hybrid neuro-Symbolic approaches, they retain the modularity of Production rules and their naturalness in a large degree. In the improved approach, various types of indices are assigned to cases according to different roles they play in neurule-based reasoning, instead of one. Thus, an enhanced knowledge representation scheme is derived resulting in accuracy improvement. Experimental results demonstrate its effectiveness.
-
Incrementally updating a hybrid rule base based on empirical data
Expert Systems, 2007Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. One way that the neu rules can be produced is from training examples/patterns, extracted from empirical data. H owever, in certain application fields not all of the training examples are available a priori. A number of them become available over time. In those cases, updating a neurule base is necessary. In this paper , methods for updating a hybrid rule base, consisting of neurules, to reflect the availability of new training examples are presented. They can b e considered as a type of incremental learning method s that retain the entire induced hypothesis and all past training examples. The methods are efficient, since they require the least possible retraining ef fort and the number of the produced neurules is kept as small as possible. Experimental results that prove the above argument are presented.
-
ICTAI - Updating a hybrid rule base with new empirical source knowledge
14th IEEE International Conference on Tools with Artificial Intelligence 2002. (ICTAI 2002). Proceedings., 2002Co-Authors: Jim Prentzas, Ioannis Hatzilygeroudis, Athanasios K. TsakalidisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. Each neurule is represented as an adaline unit. One way that the neurules can he produced is from training examples (empirical source knowledge). However, in certain application fields not all of the training examples are available a priori. A number of them become available over time. In these cases, updating the corresponding neurules is necessary. In this paper, methods for updating a hybrid rule base, consisting of neurules, to reflect the availability of new training examples are presented The methods are efficient, since they require the least possible retraining effort and the number of the produced neurules is kept as small as possible.
-
ECCBR - Integrating Hybrid Rule-Based with Case-Based Reasoning
Lecture Notes in Computer Science, 2002Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:In this paper, we present an approach integrating neurule-based and case-based reasoning. Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist representation (adaline unit). Each neurule is represented as an adaline unit. One way that the neurules can be produced is from Symbolic rules by merging the Symbolic rules having the same conclusion. In this way, the number of rules in the rule base is decreased. If the Symbolic rules, acting as source knowledge of the neurules, do not cover the full complexities of the domain, accuracy of the produced neurules is affected as well. To improve accuracy, neurules can be integrated with cases representing their exceptions. The integration approach enhances a previous method integrating Symbolic rules with cases. The use of neurules instead of Symbolic rules improves the efficiency of the inference mechanism and allows for drawing conclusions even if some of the inputs are unknown.
-
ECAI - Updating a hybrid rule base with changes to its Symbolic source knowledge
2002Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. One way that neurules (target knowledge) can be produced is by converting Symbolic rules (source knowledge). However, source knowledge may change, so that updating corresponding target knowledge is necessary. Changes concern insertion of new and removal of old Symbolic rules. In this paper, methods for updating target knowledge to follow changes made in corresponding source knowledge are presented. The methods are efficient in the sense that they do not require retraining of the whole affected part of the target knowledge, but of as small portion of it as possible.
Jim Prentzas - One of the best experts on this subject based on the ideXlab platform.
-
SETN - Improving the Integration of Neuro-Symbolic Rules with Case-Based Reasoning
Artificial Intelligence: Theories Models and Applications, 2008Co-Authors: Jim Prentzas, Ioannis Hatzilygeroudis, Othon MichailAbstract:In this paper, we present an improved approach integrating rules, neural networks and cases, compared to a previous one. The main approach integrates neurules and cases. Neurules are a kind of integrated rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. Each neurule is represented as an adaline unit. The main characteristics of neurules are that they improve the performance of Symbolic rules and, in contrast to other hybrid neuro-Symbolic approaches, they retain the modularity of Production rules and their naturalness in a large degree. In the improved approach, various types of indices are assigned to cases according to different roles they play in neurule-based reasoning, instead of one. Thus, an enhanced knowledge representation scheme is derived resulting in accuracy improvement. Experimental results demonstrate its effectiveness.
-
Incrementally updating a hybrid rule base based on empirical data
Expert Systems, 2007Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. One way that the neu rules can be produced is from training examples/patterns, extracted from empirical data. H owever, in certain application fields not all of the training examples are available a priori. A number of them become available over time. In those cases, updating a neurule base is necessary. In this paper , methods for updating a hybrid rule base, consisting of neurules, to reflect the availability of new training examples are presented. They can b e considered as a type of incremental learning method s that retain the entire induced hypothesis and all past training examples. The methods are efficient, since they require the least possible retraining ef fort and the number of the produced neurules is kept as small as possible. Experimental results that prove the above argument are presented.
-
ICTAI - Updating a hybrid rule base with new empirical source knowledge
14th IEEE International Conference on Tools with Artificial Intelligence 2002. (ICTAI 2002). Proceedings., 2002Co-Authors: Jim Prentzas, Ioannis Hatzilygeroudis, Athanasios K. TsakalidisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. Each neurule is represented as an adaline unit. One way that the neurules can he produced is from training examples (empirical source knowledge). However, in certain application fields not all of the training examples are available a priori. A number of them become available over time. In these cases, updating the corresponding neurules is necessary. In this paper, methods for updating a hybrid rule base, consisting of neurules, to reflect the availability of new training examples are presented The methods are efficient, since they require the least possible retraining effort and the number of the produced neurules is kept as small as possible.
-
ECCBR - Integrating Hybrid Rule-Based with Case-Based Reasoning
Lecture Notes in Computer Science, 2002Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:In this paper, we present an approach integrating neurule-based and case-based reasoning. Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist representation (adaline unit). Each neurule is represented as an adaline unit. One way that the neurules can be produced is from Symbolic rules by merging the Symbolic rules having the same conclusion. In this way, the number of rules in the rule base is decreased. If the Symbolic rules, acting as source knowledge of the neurules, do not cover the full complexities of the domain, accuracy of the produced neurules is affected as well. To improve accuracy, neurules can be integrated with cases representing their exceptions. The integration approach enhances a previous method integrating Symbolic rules with cases. The use of neurules instead of Symbolic rules improves the efficiency of the inference mechanism and allows for drawing conclusions even if some of the inputs are unknown.
-
ECAI - Updating a hybrid rule base with changes to its Symbolic source knowledge
2002Co-Authors: Jim Prentzas, Ioannis HatzilygeroudisAbstract:Neurules are a kind of hybrid rules that combine a Symbolic (Production rules) and a connectionist (adaline unit) representation. One way that neurules (target knowledge) can be produced is by converting Symbolic rules (source knowledge). However, source knowledge may change, so that updating corresponding target knowledge is necessary. Changes concern insertion of new and removal of old Symbolic rules. In this paper, methods for updating target knowledge to follow changes made in corresponding source knowledge are presented. The methods are efficient in the sense that they do not require retraining of the whole affected part of the target knowledge, but of as small portion of it as possible.
Dariush Davani - One of the best experts on this subject based on the ideXlab platform.
-
SERA - Architectural Design and Development of a Hybrid Robotics Control System
2011 Ninth International Conference on Software Engineering Research Management and Applications, 2011Co-Authors: E. Avery, Dariush DavaniAbstract:The Symbolic and Sub-Symbolic Robotics Intelligence Control System is a hybrid robotic control architecture under development which is based on human behavior modeling concepts from cognitive psychology and artificial intelligence disciplines. It mixes high level Symbolic Production system processing with low level sub-Symbolic processing of sensory data. This paper will detail the internal structures and algorithms used in SS-RICS over the last seven years of development. We will also discuss the ongoing development and future of the system.
-
Architectural design and development of a hybrid robotics control system
Proceedings - 2011 9th International Conference on Software Engineering Research Management and Applications SERA 2011, 2011Co-Authors: E. Avery, Dariush DavaniAbstract:The Symbolic and Sub-Symbolic Robotics Intelligence Control System is a hybrid robotic control architecture under development which is based on human behavior modeling concepts from cognitive psychology and artificial intelligence disciplines. It mixes high level Symbolic Production system processing with low level sub-Symbolic processing of sensory data. This paper will detail the internal structures and algorithms used in SS-RICS over the last seven years of development. We will also discuss the ongoing development and future of the system. © 2011 IEEE.
Jenny Pickerill - One of the best experts on this subject based on the ideXlab platform.
-
Symbolic Production representation and contested identities
Information Communication & Society, 2009Co-Authors: Jenny PickerillAbstract:The increased flexibility facilitated by Information and Communication Technologies has enabled anti-war activists to garner great control over representations of their arguments. This paper explores the value of the Symbolic dimension of collective action through three cumulative forms of analysis: understanding how the Symbolic domain is used; explaining the strategic choices behind this use; and, finally, linking these representational choices to the subjective experience of the individual and their processes of political identity construction. Many groups lacked a coherent online strategy. The reasons for this are threefold. First, use reflects the organizational structure and ideological principles of the groups. Second, by emphasizing diversity and inclusion most groups wanted their online material to be as accessible as possible and as a result censored anything too opinionated or deemed too radical. Third, there was an enduring emphasis upon local place and face-to-face communication. This made gr...
E. Avery - One of the best experts on this subject based on the ideXlab platform.
-
SERA - Architectural Design and Development of a Hybrid Robotics Control System
2011 Ninth International Conference on Software Engineering Research Management and Applications, 2011Co-Authors: E. Avery, Dariush DavaniAbstract:The Symbolic and Sub-Symbolic Robotics Intelligence Control System is a hybrid robotic control architecture under development which is based on human behavior modeling concepts from cognitive psychology and artificial intelligence disciplines. It mixes high level Symbolic Production system processing with low level sub-Symbolic processing of sensory data. This paper will detail the internal structures and algorithms used in SS-RICS over the last seven years of development. We will also discuss the ongoing development and future of the system.
-
Architectural design and development of a hybrid robotics control system
Proceedings - 2011 9th International Conference on Software Engineering Research Management and Applications SERA 2011, 2011Co-Authors: E. Avery, Dariush DavaniAbstract:The Symbolic and Sub-Symbolic Robotics Intelligence Control System is a hybrid robotic control architecture under development which is based on human behavior modeling concepts from cognitive psychology and artificial intelligence disciplines. It mixes high level Symbolic Production system processing with low level sub-Symbolic processing of sensory data. This paper will detail the internal structures and algorithms used in SS-RICS over the last seven years of development. We will also discuss the ongoing development and future of the system. © 2011 IEEE.
