Showing 1 to 65 of 65 matching Articles
Results per page:
Export (CSV)
By
Monroy, Raúl; Bundy, Alan; Green, Ian
3 Citations
Most efforts to automate formal verification of communicating systems have centred around finitestate systems (FSSs). However, FSSs are incapable of modelling many practical communicating systems, including a novel class of problems, which we call VIPS. VIPSs are valuepassing, infinitestate, parameterised systems. Existing approaches using model checking over FSSs are insufficient for VIPSs. This is due to their inability both to reason with and about domainspecific theories, and to cope with systems having an unbounded or arbitrary state space.
We use the Calculus of Communicating Systems (CCS) (Communication and Concurrency. London: Prentice Hall, 1989) to express and specify VIPSs. We take program verification to be proving the program and its intended specification equivalent. We use the laws of CCS to conduct the verification task. This approach allows us to study communicating systems and the data such systems communicate. Automating theorem proving in this context is an extremely difficult task.
We provide automated methods for CCS analysis; they are applicable to both FSSs and VIPSs. Adding these methods to the CL^{A}M proof planner (Lecture Notes in Artificial Intelligence, Vol. 449, Springer, 1990, pp. 647, 648), we have implemented an automated verification planner capable of dealing with problems that previously required human interaction. This paper describes these methods, gives an account as to why they work, and provides a short summary of experimental results.
more …
By
Shchepin, Evgeny V.; Vakhania, Nodari N.
7 Citations
We consider the problem of scheduling of n independent jobs on m unrelated machines to minimize the max(t_{1}, t_{2},..., t_{m}), t_{i} being the completion time of machine i. In [1] was suggested a polynomial 2approximation algorithm for this problem. It was also proved that there can exist no polynomial 1.5approximation algorithm unless P = NP. Here we improve this earlier performance bound 2 to 2  1/m. In [1] is also proved a general rounding theorem, which allows to construct in polynomial time 1job approximations to the optimum, i.e. schedules with an absolute bound equal to the largest job processing time. We also improve this result and obtain (1—1/m)job approximation to optimal.
more …
By
Cairó, Osvaldo; Barreiro, Julio; Solsona, Francisco
3 Citations
Software project mistakes represent a loss of millions of dollars to thousands of companies all around the world. These software projects that somehow ran off course share a common problem: Risks became unmanageable. There are certain number of conjectures we can draw from the high failure rate: Bad management procedures, an inept manager was in charge, managers are not assessing risks, poor or inadequate methodologies where used, etc. Some of them might apply to some cases, or all, or none, is almost impossible to think in absolute terms when a software project is an ad hoc solution to a given problem. Nevertheless, there is an ongoing effort in the knowledge engineering (KE) community to isolate risk factors, and provide remedies for runaway projects, unfortunately, we are not there yet. This work aims to express some general conclusions for risk assessment of software projects, particularly, but not limited to, those involving knowledge acquisition (KA).
more …
By
Serrano, Miguel A.; Carver, Doris L.; Oca, Carlos Montes
1 Citations
We present a reengineering approach for decomposing existing objectoriented systems into subsystems that have low coupling and are suitable for distribution. We use reverse engineering techniques for the architectural and design recovery. We use objectoriented metrics techniques for the assessment of relationships and interactions between objectoriented constructs such as classes, objects, and methods. Next, we use data mining techniques to discover associations in the underlying system and clustering techniques to create a hierarchical grouping of subsystems that is convenient for guiding the allocation of the subsystems to a hierarchical network. Finally, we efficiently allocate subsystems to different sites by mapping the hierarchical decomposition of subsystems to a hierarchical network representation. For the implementation, we use middleware technologies.
more …
By
Akker, Marjan; Hoogeveen, Han; Vakhania, Nodari
3 Citations
We consider a singlemachine online scheduling problem where jobs arrive over time. A set of independent jobs has to be scheduled on a single machine. Each job becomes available at its release date, which is not known in advance, and its characteristics, i.e., processing requirement and delivery time, become known at its arrival. The objective is to minimize the time by which all jobs have been delivered. In our model preemption is not allowed, but we are allowed to restart a job, that is, the processing of a job can be broken off to have the machine available to process an urgent job, but the time already spent on processing this interrupted job is considered to be lost. We propose an online algorithm and show that its performance bound is equal to 1.5, which matches a known lower bound due to Vestjens. For the same problem without restarts the optimal worstcase bound is known to be equal to (
$$
\sqrt 5 + 1
$$
)/2 ≈ 1.61803; this is the first example of a situation in which the possibility of applying restarts reduces the worstcase performance bound, even though the processing times are known.
more …
By
Li, Hujun; Li, Fansheng; Sung, Andrew H.; Weiss, William W.
Show all (4)
1 Citations
This paper presents a novel application of fuzzy logic in the interpretation of well logs, specifically, in determining the formation rock types in Petroleum Engineering. To solve this practical problem, a new inference algorithm is proposed. The interpretation of well logs is a decisionmaking problem where the issue is to utilize (and compromise) knowledge from human experts, evidence from well logs, and information from other sources. This research was motivated by the fact that fuzzy logic has proven to be highly effective in many applications involving uncertainties. Comparing with neural networks, this fuzzylogicbased method avoids the problems of training data collection, network training, and unavailability of rules or knowledge used in the interpretation. This results in an algorithm that is effective and inexpensive.
more …
By
Lemaître, Christian; FallahSeghrouchni, Amal
1 Citations
Based on J. Habermas/Bühler’ Communicative Action theory we propose a novel framework that goes beyond the classical speech act theory and its intentionalistic interpretations. We introduce a comprehensive theory of meaning for communication acts assuming that the content of natural language utterances can be classified in three different domains of discourse, each one with a different type of semantic validation: the domain of objective facts, the internal or subjective domain of the sender, and the social relational domain of the sender and the receiver. Following Habermas, we introduce also a crucial shift in the agent interaction approach, focusing on the conversation control issues, on the receiver and not on the sender. We claim these two new approaches of mutiagent interactions will allow to control and manage the complex interactions among agents in open real world applications.
more …
By
Mariano, Carlos; Morales, Eduardo
4 Citations
Many problems can be characterized by several competing objectives. Multiple objective optimization problems have recently received considerable attention specially by the evolutionary algorithms community. Their proposals, however, require an adequate codification of the problem into strings, which is not always easy to do. This paper introduces a new algorithm, called MDQL, for multiple objective optimization problems which does not suffer from previous limitations. MDQL is based on a new distributed Qlearning algorithm, called DQL, which is also introduced in this paper. Furthermore, an extension for applying reinforcement learning to continuos functions is also given. Successful results of MDQL on a continuos non restricted problem whose Pareto front is convex and on a continuos nonconvex problem with restrictions are described.
more …
By
Ranjan, D.; Pontelli, E.; Gupta, G.; Longpre, L.
Show all (4)
9 Citations
Abstract.
In this paper we analyze the complexity of the Temporal Precedence Problem on pointer machines. The problem is to support efficiently two operations: insert and precedes. The operation insert(a) introduces a new element a , while precedes(a,b) returns true iff element a was temporally inserted before element b . We provide a solution to the problem with worstcase time complexity O( lg lg n) per operation, where n is the number of elements inserted. We also demonstrate that the problem has a lower bound of Ω( lg lg n) on pointer machines. Thus the proposed scheme is optimal on pointer machines.
more …
By
Palacios Pérez, José Juan
Interaction among agents is the distinguish property of MultiAgent Systems (MAS). In order to take advantage of such interaction w.r.t goals (both local for each agent and common for a group of agents) distributed planning, (i.e. the generation of coordination and cooperation of activities) is fundamental. Distributed MAS planning can be uniformly modeled in terms of resources: formal plan representation and effective interacting methods. Respectively, plan representation can be expressed by means of logical proofs and interacting methods by means of asynchronous communication. This paper presents preliminary considerations for the definition of a model for distributed MAS planning based on resources. For this mean, HACL (Higher Order Asynchronous Communications in Linear Logic) is used for analysis on the structural model RPN (Recursive Petri Nets) for MultiAgent distributed planning and for specification of MAS interaction protocols.
more …
By
Guzmán, Adolfo; Olivares, Jesús; Demetrio, Araceli; Domínguez, Carmen
Show all (4)
4 Citations
At CIC we have developed a model that enables multithreaded agents that do not share the same ontology, to interact and interchange information among them.
The behavior of each agent is defined in a highlevel language with the following features:
1
Each agent and each interaction can be described by several sequences of instructions that can be executed concurrently. Some threads belong to an agent, others are inherited from the scripts which they play or perform.
2
Of all the threads, the agent must select which ones to execute, perhaps choosing between contradictory or incompatible threads.
3
The model allows communications between agents having different data dictionaries (ontologies), thus requiring conversion or matching among the primitives they use (§4).
4
Some of the threads can be partially executed, thus giving rise to the idea of a “degree of satisfaction” (§6.2.1).
5
The world on which the agents thrive suffers unexpected events (§3), to which some agents must react, throwing them out of their current behavior(s).
The model, language, executing environment and interpreter are described. Some simple examples are presented. The model will be validated using test cases based on real situations like electronic commerce, product delivery [including embedding agents in hardware], and automatic filling of databases (§6.2.2).
more …
By
Osorio, Mauricio; Nieves, Juan Carlos; Zacarias, Fernando; Saucedo, Erika
Show all (4)
We introduce the new paradigm of HighLevel NonMonotonic reasoning (HLNM). This paradigm is the consolidation of our recent results on disjunctions, sets, explicit and implicit negation, and partialorder clauses. We show how these concepts are integrated in a natural way into the standard logic programming framework. For this purpose, we present several well known examples from the literature that motivate the need of this new paradigm. Finally, we define a declarative semantics for HLNM reasoning.
more …
By
Navarrete, Dulcinea; Dávila, Rogelio; Sánchez, Alfredo
In this paper, SAVIA is introduced; a system for attaining automatic translation (EnglishtoSpanish) of morphological descriptions of plants. It was developed at Universidad de las Américas Puebla, as part of a collaborative research with the Flora of North America (FNA) project, currently under construction at the Missouri Botanical Garden [Schnase et al. 1994]. The system has been put forward as a potential solution to the problem of knowledge distribution and multilingual access to Digital Libraries.
Digital Libraries (DLs) are one of the emerging fields of study in computer science and one of the more interesting multidisciplinary research areas. Knowledge distribution is one of its main objectives; hence, the need to develop multilingual applications that make knowledge available to every community that cannot communicate in the language in which DLs material and services where implemented.
A prototype of the system has been developed using technologies based on GlossaryBased Machine Translation and the semantic grammar model. Some examples are presented showing the possibilities of the system. Conclusions will be drawn from the performance of the system and suggestions are included considering further developments.
more …
By
SánchezAnte, Gildardo; Ramos, Fernando; Frausto, Juan
5 Citations
This paper presents some preliminary results on the application of a Cooperative Simulated Annealing algorithm (COSA) for solving the problem of path planning in MultiRobot systems. The main idea is to generate paths for each robot in the system without taking care of the other robots, and then coordinate the paths for all the robots. Paths are generated using a variation of the probabilistic roadmaps (PRM), and the coordination of the paths is achieved with a Cooperative Simulated Annealing (COSA). To evaluate the system, several experiments for environments constituted by two planar robots with up to five dof and various convex obstacles were run. The results obtained aim to continue working on this line.
more …
By
Gelbukh, Alexander F.
5 Citations
A fulltext information retrieval system has to deal with various phenomena of string equivalence: ignore case matching, morphological inflection, derivation, synonymy, and hyponymy or hyperonymy. Technically, this can be handled either at the time of indexing by reducing equivalent strings to a common form or at the time of query processing by enriching the query with the whole set of the equivalent forms. We argue for that the latter way allows for greater flexibility and easier maintenance, while being more affordable than it is usually considered. Our proposal consists in enriching the query only with those forms that really appear in the document base. Our experiments with a thesaurusbased information retrieval system showed only insignificant increase of the query size on average with a 200megabyte document base, even with highly inflective Spanish language.
more …
By
Pèrez, Joaquín; Pazos, Rodolfo; Frausto, Juan; Romero, David; Cruz, Laura
Show all (5)
6 Citations
This paper presents an extension of the DFAR mathematical optimization model, which unifies the fragmentation, allocation and dynamical migration of data in distributed database systems. The extension consists of the addition of a constraint that models the storage capacity of network sites. This aspect is particularly important in large databases, which exceed the capacity of one or more sites. The Threshold Accepting Algorithm is a variation of the heuristic method known as Simulated Annealing, and it is used for solving the DFAR model. The paper includes experimental results obtained for large test cases.
more …
By
Monroy, Raúl; Bundy, Alan; Green, Ian
Unique Fixpoint Induction, UFI, is a chief inference rule to prove the equivalence of recursive processes in CCS [7]. It plays a major role in the equational approach to verification. This approach is of special interest as it offers theoretical advantages in the analysis of systems that communicate values, have infinite state space or show parameterised behaviour.
The use of UFI, however, has been neglected, because automating theorem proving in this context is an extremely difficult task. The key problem with guiding the use of this rule is that we need to know fully the state space of the processes under consideration. Unfortunately, this is not always possible, because these processes may contain recursive symbols, parameters, and so on.
We introduce a method to automate the use of UFI. The method uses middleout reasoning and, so, is able to apply the rule even without elaborating the details of the application. The method introduces variables to represent those bits of the processes’ state space that, at application time, were not known, hence, changing from equation verification to equation solving.
Adding this method to the equation plan developed by Monroy, Bundy and Green [8], we have implemented an automated verification planner. This planner increases the number of verification problems that can be dealt with fully automatically, thus improving upon the current degree of automation in the field.
more …
By
RodríguezVázquez, Katya; Fleming, Peter J.
1 Citations
This paper demonstrates how genetic programming can be used for solving problems in the field of nonlinear system identification of rational models. By using a twotree structure rather than introducing the division operator in the function set, this genetic programming approach is able to determine the “true” model structure of the system under investigation. However, unlike use of the polynomial, which is linear in the parameters, use of rational model is nonlinear in the parameters and thus noise terms cannot be estimated properly. By means of a second optimisation process (realcoded GA) which has the aim of tunning the coefficients to the “true” values, these parameters are then correctly computed. This approach is based upon the wellknown NARMAX model representation, widely used in nonlinear system identification.
more …
By
Vallejo, Edgar E.; Ramos, Fernando
2 Citations
In this work, genetic programming systems are viewed as cooperative processes: individuals in the population cooperate to evolve a global solution. There are two basic forms of cooperation for distributed problem solving: tasksharing and resultsharing. We introduce two models that enable cooperation in genetic programming systems. The first model is based on tasksharing and the second one is based on resultsharing. We examine the effects of both forms of cooperation on the performance of genetic programming systems solving the insect locomotion problem. This article demostrates that cooperative genetic programming can be used to evolve several commonly observed gaits in insects.
more …
By
Sanvicente Sánchez, Héctor; Frausto Solís, Juan
1 Citations
Simulated Annealing (SA) is usually implemented in a sequential way. We propose a Methodology to Parallel the Simulated Annealing Algorithm (MPSA). This methodology carries out the parallelization of the cycle that controls the temperature in the algorithm. This approach lets a massive parallelization. The initial solution for each internal cycle may be set through a Monte Carlo random sampling to adjust the Boltzmann distribution at the cycle beginning. In MPSA the communication scheme and its implementation must be in an asynchronous way. Through a theoretical analysis we establish that any implementation of MPSA leads to a Simulated Annealing Parallel Algorithm (SAPA) that is in general more efficient than its sequential implementation version.
more …
By
Vakhania, Nodari
1 Citations
Classical Job Shop Scheduling problem describes operation in several industries. In this classical model, there is only one machine available for each group of tasks, and the precedence relations between the tasks are restricted to serialparallel. Here we suggest a generaliza tion, which allows parallel unrelated machines and arbitrary precedence relations between the tasks. The feasible solution space of the generali zed problem (which is significantly larger than that of a corresponding version without parallel machines) is efficiently reduced to an essenti ally smaller subset. The use of this reduced subset, instead of the whole feasible set, is beneficial. We propose global and local search algorithms which start from the reduced solution set.
more …
By
Gelbukh, Alexander; Sidorov, Grigori; Bolshakov, Igor A.
4 Citations
Resolution of referential ambiguity is one of the most challenging problems of natural language processing. Especially frequently it is faced within dialogues. We present a heuristic algorithm for detection of the indirect antecedents for dialogue phrases based on the use of a dictionary of prototypic scenarios associated with each headword as well as of a thesaurus of the standard type. The conditions for filtration of the candidates for the antecedent are presented. We also present a similar algorithm for reconstruction of elliptical phrases of a special kind using a combinatory dictionary.
more …
By
Ibarra Rivera, Miguel Angel; Favela Vara, Jesús; LópezLópez, Aurelio
The availability of a well structured lexicon and of a parser that takes advantage of that structure for the analysis of sentences are of fundamental importance in a natural language processing system. A syntacticconceptual analyzer of sentences in the Spanish language based on the cognitive model of Kavi Mahesh is presented. A Spanish lexicon extracted from WordNet is used for disambiguation. The terms in the lexicon are restricted to those used in the gardening domain. The structure defining each concept was widened, including for each one of the concepts three additional attributes: the actions that the entity defined by the concept usually accomplishes, the actions that others accomplish on that entity, and the actions that the entity accomplishes on itself. The analyzer itself is constituted by a finite state automata, of the augmented transition network kind, and represents a part of the Spanish grammar with which the disambiguation method is illustrated. The procedure correctly desambiguates the sentences, though is sensitive to the placement of the associated actions in the concepts within the hierarchy of the lexicon. Based on a wellstructured lexicon, the method operates correctly.
more …
By
Decouchant, Dominique; MartínezEnríquez, Ana María
5 Citations
Related to the cooperative editing research domain and to the works currently developed on the World Wide Web, we present AllianceWeb, an editing and cooperative authoring system on the Web. Using this system, authors distributed among the world can cooperate producing large documentations in a consistent and concerted way. Taking benefits of the design and experiment of the Alliance cooperative editor on Internet, the AllianceWeb approach proposes a mixed architecture (hybrid and/or fully distributed) for the document storage and access. These two architectures are integrated to provide a concerted, secure and parameterizable cooperative editing support.
Following that, we highlight the main aspects of the group awareness function which allows each author to diffuse his contribution to other coauthors, and to control the way by which other contributions are integrated in his environment. In order to support this function, fundamental for every groupware application, techniques of Artificial Intelligence research domain allow to design and to define a selfadaptive cooperative interaction environment, parametrized by user preferences. Thus, the problem and the characteristics of a group awareness inference engine are defined.
more …
By
Rodríguez, Andrés F.; Vadera, Sunil; Sucar, L. Enrique
An exemplarbased model with foundations in Bayesian networks is described. The proposed model utilises two Bayesian networks: one for indexing of categories, and another for identifying exemplars within categories. Learning is incrementally conducted each time a new case is classified. The representation structure dynamically changes each time a new case is classified and a prototypicality function is used as a basis for selecting suitable exemplars. The results of evaluating the model on three datasets are presented.
more …
By
Romero, Leonardo; Morales, Eduardo; Sucar, Enrique
3 Citations
A mobile robot must explore its workspace in order to learn a map of its environment. Given the perceptual limitations and accuracy of its sensors, the robot has to stay close to obstacles in order to track its position and never get lost. This paper describes a new method for exploring and navigating autonomously in indoor environments. It merges a local strategy, similar to a wall following strategy to keep the robot close to obstacles, within a global search frame, based on a dynamic programming algorithm. This hybrid approach takes advantages of local strategies that consider perceptual limitations of sensors without losing the completeness of a global search. These methods for exploring and navigating are tested using a mobile robot simulator with very good results
more …
By
Garza, Luis E.; Cantú, Francisco; Acevedo, Salvador
We propose a methodology to diagnose multiple faults in complex systems. The approach is based on the Independent Choice Logic (ICL) and comprises two phases. In phase 1 we generate the explanations of the observed symptoms and handle the combinatorial explosion with a heuristic. In phase 2 we observe process signals to detect abnormal behavior that can lead us to identify the real faulted components. A proposal is made to automate this task with Dynamic Bayesian Networks (DBNs) embedded in the ICL formalism. The overall scheme is intended to give a definite diagnosis. ICL is a framework, which comprises a theory and a development environment. We show that ICL can be scaledup to realworld, industrialstrength problems by using it in diagnosing faults in an electrical power transmission network .
more …
By
TorresJimenez, Jose; RodriguezTello, Eduardo
3 Citations
The Bandwidth Minimization Problem for Graphs (BMPG) can be defined as finding a labeling for the vertices of a graph, where the maximum absolute difference between labels of each pair of connected vertices is minimum. The most used measure for the BMPG algorithms isβ, that indicates only the maximum of all absolute differences.
After analyzing some drawbacks of β, a measure, calledγ, which uses a positional numerical system with variable base and takes into account all the absolute differences of a graph is given.
In order to test the performance of γ and β a stochastic search procedure based on a Simulated Annealing (SA) algorithm has been applied to solve the BMPG. The experiments show that the SA that uses γ has better results for many classes of graphs than the one that uses β.
more …
By
Ayala, Gerardo; Hernández, Arlette
1 Citations
In this paper we propose the software components and functionality needed for the implementation of an environment that supports lifelong learning. The paper presents the requirements and design issues of intelligent agents that support lifelong learning in a community. A user agent assists the learner in the construction of her/his learning plan, and the configuration of discussion groups. An information agent is modeled in order to assist the learner in searching the location of information and knowledge considered relevant. A facilitator agent is discussed, designed for supporting the social construction of knowledge in the community. The paper covers the proposal for learner modeling and the construction of learning plans, needed for this kind of learning environments.
more …
By
Gómez, Giovani; Ramos, Fernando
A cooperative strategy for solving conflicting situations on mutirobotics motion planning is presented. The strategy is performed with a function with minimal sharing information, which controls that such movements really assists to both robots. We show that a simple function with cooperative sense can retrieve good paths in conflicting situations. As a rule, finest movements are neccessary to avoid collisions in hard environments such as inside a tunnel. We developed a novel technique inspirated in regularisation and Markovian processes. This technique is currently implemented in a randomized algorithm with a cooperative strategy for solving conflicting situations that often arrive in multirobot manipulator environments. Randomized Algorithms (RA) have proven their capacity to escape from local minima. On the other hand, if collaborative attitudes are carriedout by the participants in the task, the time to solve conflicting situations can be considerably reduced. Moreover, when conflicting situations take long time to be solved, high risk of exploring not good free spaces can arrive and hence the search does not often drive directly the robots to their goals. Based on the above considerations, we built a framework of cooperative RA’s aiming to facilitate the coordination of robots by establishing commitments to cooperate when conflicting situations appear, and by interacting through the exchange of data to go as directly as possible towards their respective goals.
more …
By
MontesyGómez, Manuel; Gelbukh, Alexander; LópezLópez, Aurelio
15 Citations
In intelligent knowledgebased systems, the task of approximate matching of knowledge elements has crucial importance. We present the algorithm of comparison of knowledge elements represented with conceptual graphs. The method is based on wellknown strategies of text comparison, such as Dice coefficient, with new elements introduced due to the bipartite nature of the conceptual graphs. Examples of comparison of two pieces of knowledge are presented. The method can be used in both semantic processing in natural language interfaces and for reasoning with approximate associations.
more …
By
Ramirez, Carlos
This paper presents a demonstration of a Theory for Episodic Learning applied to Information Retrieval. The theory is compatible with CaseBased Reasoning and is used to model how contextual information about users, their aims and searchsessions experience may be represented in case bases, and automatically tuned up. The proposed model shows how query processing may be extended by deriving information from the case bases, taking advantage of user profiles and a history of queries derived from searchsessions. The paper includes a demonstration of the program IRBOC, presenting a fully worked example.
more …
By
Mariano, Carlos; Morales, Eduardo
7 Citations
This paper describes a new algorithm, called MDQL, for the solution of multiple objective optimization problems. MDQL is based on a new distributed Qlearning algorithm, called DQL, which is also introduced in this paper. In DQL a family of independent agents, explo ring different options, finds a common policy in a common environment. Information about action goodness is transmitted using traces over state action pairs. MDQL extends this idea to multiple objectives, assigning a family of agents for each objective involved. A nondominant criterion is used to construct Pareto fronts and by delaying adjustments on the rewards MDQL achieves better distributions of solutions. Furthermore, an extension for applying reinforcement learning to continuous functions is also given. Successful results of MDQL on several testbed problems suggested in the literature are described.
more …
By
GarcíaRodríguez, Armando; RodríguezDagnino, Ramón M.; Douligeris, Christos
This paper presents a dynamic bandwidth allocation system for realtime variable bit rate (VBR) video transport in asynchronous Transfer mode (ATM) networks. This system takes advantage of scene changes in the video trace on a scale larger than a second, and it adapts the bandwidth as needed. An improvement on efficiency is achieved by assigning bandwidth for the transport of VBR video and having more than one predictor in parallel with different prediction horizons, hence this scheme reduces the processing time for the bandwidth adaptation with no significant degradations on the queue statistics. The link capacity required for a specific session is a function of the input traffic, which is characterized by its spectral characteristic. In particular, we use the low frequency band of the power spectrum, which is extracted from the stochastic input by a low pass filter. The predictor is a neural network (NN) called ”PiSigma Network”, and the output of this predictor is interpreted according to the prediction horizon in use.
more …
By
Aguera, Ana S.; Guerra, Alejandro; Martínez, Manuel
1 Citations
The main purpose of this work is to explore the application of Memory Based Reasoning (MBR) to adaptive behavior in agents. We discuss the design of an interface agent for email management assistance as a prototype for the experimental assesment of an MBR learning algorithm performance. Results are discussed and a brief summary of conclusions follows, as well as a sketch of future work to be done.
more …
By
Bourguet, Rafael E.; Soto, Rogelio
1 Citations
This paper addresses the first step about collecting qualitative knowledge from persons working for learning organizations. Mental models of policymakers are considered the sources of knowledge. The main idea is to represent their qualitative knowledge into a computer simulation model. System Dynamics methodology is used. Results of the process carried out with a team of two consultants on a hotel system are presented. Conclusions are derived and future research is discussed to construct knowledgebased systems that incorporate simulation models for learning purposes in business organizations.
more …
By
Rayón Villela, P.; Sossa Azuela, J. H.
2 Citations
The Adaptive Resonance Theory ART2 [1] is used as a non supervised tool to generate clusters. The clusters generated by an ART2 Neural Network (ART2 NN), depend on a vigilance threshold (ρ). If ρ is near to zero, then a lot of clusters will be generated; if ρ is greater then more clusters will be generated. To get a good performance, this ρ has to be suitable selected for each problem. Until now, no technique had been proposed to automatically select a proper ρ for a specific problem. In this paper we present a first way to automatically obtain the value of ρ, we also illustrate how it can be used in supervised and unsupervised learning. The goal to select a suitable threshold is to reach a better performance at the moment of classification. To improve classification, we also propose to use a set of feature vectors instead of only one to describe the objects. We present some results in the case of character recognition.
more …
By
Maya, Selene; Reynoso, Rocio; Torres, César; AriasEstrada, Miguel
Show all (4)
7 Citations
An FPGA based Artificial Neural Network is proposed. The neuron is based on a spiking scheme where signals are encoded in a stochastic pulse train. The neuron is composed of a synaptic module and a summingactivation module. The architecture of the neuron is characterized and its FPGA implementation is presented. The basic spiking neuron is used to implement a basic neural network. An extension of the neuron architecture to include an addressevent protocol for signal multiplexing in a single line is proposed. VHDL simulations and FPGA synthesis results are discussed.
more …
By
MontesyGómez, Manuel; LópezLópez, Aurelio; Gelbukh, Alexander
18 Citations
The use of conceptual graphs for the representation of text contents in information retrieval is discussed. A method for measuring the similarity between two texts represented as conceptual graphs is presented. The method is based on wellknown strategics of text comparison, such as Dice coefficient, with new elements introduced due to the bipartite nature of the conceptual graphs. Examples of the representation and comparison of the phrases are given. The structure of an information retrieval system using twolevel document representation, traditional keywords and conceptual graphs, is presented.
more …
By
Kirschning, Ingrid; Aguas, Nancy
This paper presents a new method for the verification of the correct pronunciation of spoken words. This process is based on speech recognition technology. It can be particularly useful when applied to the field of SLA (Second Language Acquisition) in learning environments or ComputerAided Language Learning (CALL) systems, where the students can practice their pronunciation skills. This method uses an artificial neural network plus a specific grammar for each utterance to compare the text of the expected utterance with the sequence of phonemes recognized in the speech input, in order to detect the pronunciation errors.
more …
By
Weitzenfeld, Alfredo; Peguero, Oscar; Gutiérrez, Sebastián
2 Citations
As neural systems become large and complex, sophisticated tools are needed to support effective model development and efficient simulation processing. Initially, during model development, rich graphical interfaces linked to powerful programming languages and component libraries are the primary requirement. Later, during model simulation, processing efficiency is the primary concern. Workstations and personal computers are quite effective during model development, while parallel and distributed computation become necessary during simulation processing. We give first an overview of modeling and simulation in NSL together with a depth perception model example. We then discuss current and future work with the NSL/ASL system in the development and simulation of modular neural systems executed in a single computer or distributed computer network.
more …
By
Akiyama, J.; Kaneko, A.; Kano, M.; Nakamura, G.; RiveraCampo, E.; Tokunaga, S.; Urrutia, J.
Show all (7)
4 Citations
In this paper we study the following problem: how to divide a cake among the children attending a birthday party such that all the children get the same amount of cake and the same amount of icing. This leads us to the study of the following. A perfect kpartitioning of a convex set S is a partitioning of S into k convex pieces such that each piece has the same area and
$\frac{1}{k}$
of the perimeter of S . We show that for any k, any convex set admits a perfect kpartitioning. Perfect partitionings with additional constraints are also studied.
more …
By
Buen Rodriguez, Pablo R.; Morales, Eduardo F.; Vadera, Sunil
An algorithm developed to help an expert generate rules is presented. The algorithm, which has been called RuLess, consists of two main stages: (i) a session to incrementally capture the rules, and (ii) a mechanism to simplify the rules. In general, it can be used on environments where there is a finite set of possible examples which are not available in advance, and from which a set of classification rules needs to be produced. It is useful in domains in which all the attributes are discrete and the number of examples is not too large, as the user needs to manually classify all the examples. RuLess was used to generate the tutoring rules of LacePro, a multifunctional system to learn, apply and consult established procedures. Based on the rules obtained for LacePro, the RuLess method was compared against the CN2 and Ripple Down Rules methods, which are two wellknown rule generation procedures.
more …
By
UribeGutierrez, Sergio; MartinezAlfaro, Horacio
3 Citations
This paper describes the working philosophy of the behaviorbased architecture applied to the design of mobile robots. The main advantages with respect to the conventional architecture are presented together with the pioneering ideas of its creator. Finally I present the case of three different computer beings (softbots); they were programmed with the behaviorbased architecture with different levels of “intelligence”.
more …
By
Ibargüengoytia, Pablo H.; Sucar, L. Enrique; Morales, Eduardo
2 Citations
Diagnosis, in artificial intelligence, has traditionally utilized heuristic rules which in many domains are difficult to acquire. An alternative approach, modelbased diagnosis, utilizes a model of the system and compares its predicted behavior against the actual behavior of the system for diagnosis. This paper presents a novel technique based on probabilistic models. Therefore, it is natural to include uncertainty in the model and in the measurements for diagnosis. This characteristic makes the proposed approach suitable for applications where reliable measurements are unlikely to occur or where a deterministic analytical model is difficult to obtain. The proposed approach can detect single or multiple faults through a vector of probabilities which reflects the degree of belief in the state of all the components of the system. A comparison against GDE, a classical approach for multiple fault diagnosis, is given.
more …
By
Coello Coello, Carlos A.; Zavala, Rosa Laura G.; García, Benito Mendoza; Aguirre, Arturo Hernández
Show all (4)
14 Citations
In this paper we propose an application of the Ant System (AS) to optimize combinational logic circuits at the gate level. We define a measure of quality improvement in partially built circuits to compute the distances required by the AS and we consider as optimal those solutions that represent functional circuits with a minimum amount of gates. The proposed methodology is described together with some examples taken from the literature that illustrate the feasibility of the approach.
more …
By
Romero, Leonardo; Morales, Eduardo; Sucar, Enrique
4 Citations
A new method for learning probabilistic gridbased maps of the environment of a mobile robot is described. New contributions on the three major components of map learning, namely, sensor data fusion, exploration and position tracking, are proposed. In particular, new models of sensors and a way of sensor data fusion that takes advantage of multiple viewpoints are presented. A new approach to control the exploration of the environment taking advantages of local strategies but without losing the completeness of a global search is given. Furthermore, a robot position tracking algorithm, based on polar and rectangular correlations between laser range data and the map is also introduced. Experimental results for the proposed approach using a mobile robot simulator with odometer, ultrasonic and laser range sensors (implemented with laser pointers and a camera), moving in an indoor environment, are described.
more …
By
ArroyoFigueroa, G.; Sucar, L. Enrique
A methodology for online diagnosis and prediction of power plant disturbances has been developed, implemented, and tested. The approach is sufficiently comprehensive to enable a wide variety of disturbances to be analyzed correctly and efficiently. The analysis is based on a novel knowledge representation, called Temporal Nodes Bayesian Networks (TNBN), a type of probabilistic network that include temporal information. A TNBN has a set of temporal nodes that represent state changes. Each temporal node is defined by a event and a time interval associated to its occurrence. The method has been implemented and integrated with a power plant training simulator. Disturbance models for the feedwater and superheater systems have been developed and implemented as the knowledge database for the disturbance analysis system.
more …
By
ZozayaGorostiza, Carlos; OrellanaMoyao, David R.
This article describes several modifications performed to the credit assignment mechanism of Goldberg’s Simple Classifier System [4] and the results obtained when solving a problem that requires the formation of classifier chains. The first set of these modifications included changes to the formula used to compute the effective bid of a classifier by taking into consideration the reputation of the classifier and the maximum bid of the previous auction in which a classifier was active. Noise was made proportional to the strength of the classifier and specificity was incorporated as an additional term in the formula that is independent from the bid coefficient. A second set of changes was related to the manner in which classifiers belonging to a chain may receive a payoff or a penalty from the environment in addition to the payments obtained from succeeding classifiers. We also tested the effect that bridge classifiers [13] have in the solution of the example problem by allowing the creation of shorter chains. Some experiments in which classifiers were better informed gave better results than those in which only the noise and the specificity were included in the computation of the effective bid.
more …
By
Esposito, Anna; Ezin, Eugène C.; ReyesGarcia, Carlos A.
Noise canceling is an adaptive interference filtering technique that has shown to be highly adventageous in many applications where fixed filters are not efficient. We present an experimental neurofuzzy inference system, based on the ANFIS architecture, which has been implemented with the objective to perform nonlinear adaptive noise cancellation from speech. The novelty of the system described in the present paper, with respect to our previus work, consists in a different set up, which requires two inputs with seven membership functions each, and uses a second order sinc function to generate the nonlinear distortion of the noise. This set up allows a better generalization to the system for learning the noise features. Indeed, the system was trained only once during few epochs, with a sample of babble noise, but it was able to clean speech sentences corrupted not only with the same noise, but also with car, traffic, and white noise. The average improvement, in terms of SNR, was 37 dB without further training, resulting in a great reduction of the computational time.
more …
By
Weitzenfeld, Alfredo
1 Citations
The study of biological systems has inspired the development of a large number of neural network architectures and robotic implementations. Through both experimentation and simulation biological systems provides a means to understand the underlying mechanisms in living organisms while inspiring the development of robotic applications. Experimentation, in the form of data gathering (ethological physiological and anatomical), provides the underlying data for simulation generating predictions to be validated by theoretical models. These models provide the understanding for the underlying neural dynamics, and serve as basis for simulation and robotic experimentation. Due to the inherent complexity of these systems, a multilevel analysis approach is required where biological, theoretic and robotic systems are studied at different levels of granularity. The work presented here overviews our existing modeling approach and describes current simulation results.
more …
By
Leal Ascencio, Raúl; Perez Cisneros, Marco
1 Citations
Artificial Neural Networks (ANN) are an emerging technology, yet, in continuous dynamic behavior, much work has been done to attempt to generate a formal method to design a controller based on this technology. Based on this fact we are starting to work towards contributing to the generation of a formal method of the application of neurocontrol and thus present several control schemes using multilayer perceptrons and also Albus’ Cerebellar Model Articulation Controller (CMAC). The control schemes are presented and explained. The direct inverse neurocontroller model is used as the system controller without a linear controller. Results from two neurocontrollers are presented in a comparative way versus linear system using a 2DOF plant.
more …
By
Flores, M.; Molina, A.
7 Citations
The world is living a fast changing process, where the creation of strategic alliances is one important result of this continuous change. A new way of doing business where companies can share their core competencies in order to compete in this global and changing economy is the creation of Virtual Enterprises. To enable the formation of Virtual Enterprises, Virtual Industry Clusters (VIC) should be created. After market opportunities are identified, global partners will need to be searched within Virtual Industry Clusters for specific products, business processes or technologies in order to create Virtual Enterprises. In this paper a Methodology for the Creation of Virtual Industry Clusters is proposed and a case study is presented to describe how the methodology has been applied to create VIRPLAS — A Virtual Industry Cluster of Mexican Plastics companies.
more …
By
Baruch, Ieroham S.; Martín Flores, J.; Carlos Martínez, J.; Nenkova, Boyka
Show all (4)
2 Citations
A twolayer Recurrent Neural Network Model (RNNM) and an improved Backpropagationthroughtime method of its learning are described. For a complex nonlinear plants identification, a fuzzyneural multimodel, is proposed. The proposed fuzzyneural model, containing two RNNMs is applied for realtime identification of nonlinear mechanical system. The simulation and experimental results confirm the RNNM applicability.
more …
By
Osorio, Mauricio; Zacarias, Fernando
3 Citations
We introduce the paradigm of HighLevel Logic Programming. This paradigm is the consolidation of our recent results on disjunctions, sets, partialorder clauses and aggregation. We show how these concepts are integrated in a natural way into the standard logic programming framework. For this purpose, we present several well known examples from the literature that support this claim. Our approach to define the declarative semantics of HLL (HighLevel Logic) programs consists on a translation of them to datalog disjunctive programs and then to use D1WFSCOMP.
more …
By
Rentería, C.; TapiaRecillas, H.
The ainvariant, the defining ideal, the dimension and the minimal distance of some ReedMuller type codes arising from the Veronese variety over a finite field are determined. Some examples are provided to illustrate the main results. These codes are a natural generalization of the projective ReedMuller codes.
more …
By
Urías, Jesús
1 Citations
Large families of permutations and a generator of pseudorandom sequences of binary words are implemented in the form of a single square array of XOR gates following the local rule of an expansive cellular automaton.
By
BayroCorrochano, Eduardo; Daniilidis, Kostas; Sommer, Gerald
42 Citations
In this paper we apply the Clifford geometric algebra for solving problems of visually guided robotics. In particular, using the algebra of motors we model the 3D rigid motion transformation of points, lines and planes useful for computer vision and robotics. The effectiveness of the Clifford algebra representation is illustrated by the example of the handeye calibration. It is shown that the problem of the handeye calibration is equivalent to the estimation of motion of lines. The authors developed a new linear algorithm which estimates simultaneously translation and rotation as components of rigid motion.
more …
By
BayroCorrochano, Eduardo; Zhang, Yiwen
19 Citations
In this paper the motor algebra for linearizing the 3D Euclidean motion of lines is used as the oretical basis for the development of a novel extended Kalman filter called the motor extended Kalman filter (MEKF). Due to its nature the MEKF can be used as online approach as opposed to batch SVD methods. The MEKF does not encounter singularities when computing the Kalman gain and it can estimate simultaneously the translation and rotation transformations. Many algorithms in the literature compute the translation and rotation transformations separately. The experimental part demonstrates that the motor extended Kalman filter is an useful approach for estimation of dynamic motion problems. We compare the MEKF with an analytical method using simulated data. We present also an application using real images of a visual guided robot manipulator; the aim of this experiment is to demonstrate how we can use the online MEKF algorithm. After the system has been calibrated, the MEKF estimates accurately the relative position of the endeffector and a 3D reference line. We believe that future vision systems being reliably calibrated will certainly make great use of the MEKF algorithm.
more …
By
Ramírez, Ana M.; García, Esther O.; Del Río, J. Antonio
35 Citations
Many aspects determine the quality of scientific journals. The impact factor is one of these quantitative parameters. However, the impact factor has a strong dependence on the journal discipline. This dependence forbids a direct comparison between different journals without introducing external considerations. In this paper, a renormalized impact factor, F_{r}, inspired in the definition of dimensionless physical parameters, is proposed. F_{r} allows a direct comparison among journals classified into different categories and, furthermore, the time evolution analysis of the journal's role in its field.
more …
By
Velasco, Fernando; Verma, Surendra P.; Guevara, Mirna
7 Citations
This work presents a comparison of relative efficiency of fourteen statistical tests to detect outliers in normally distributed samples. These tests include deviation/spread, Grubbstype, Dixontype, and highorder moment statistics. Performance for the statistical tests is evaluated using Geochemical Reference Material databases from the United States Geological Survey. Test efficiency is compared for the first application of statistics in version k = 1 and k = 2, as well as for tests in version k = 1 applied consecutively against block procedures in which k = 2, 3, or 4 values are evaluated at the same time. In both evaluations, the sensitivity of the statistical tests shows a general strong dependence on sample size. The best performance is observed for the block procedures compared to consecutive statistical tests affected by masking effects.
more …
By
Díaz De León S., J.L.; SossaAzuela, J.H.
Mathematical Morphology (MM) is a general method for image processing based on set theory. The two basic morphological operators are dilation and erosion. From these, several non linear filters have been developed, usually with polynomial complexity and this because the two basic operators depend strongly on the definition of the structural element. Most efforts to improve the algorithm's speed for each operator are based on structural element decomposition and/or efficient codification.
In this second part, the concepts developed in part I (see Díaz de León and Sossa Azuela, “Mathematical morphology based on linear combined metric spaces on Z^{1} (part I): Fast distance transforms,” Journal of Mathematical Imaging and Vision, Vol. 12, No. 2, pp. 137–154, 2000) are used to prove that it is possible to reduce the complexity of the morphological operators to zero complexity (constant time algorithms) for any regular discrete metric space and to eliminate the use of the structural element. In particular, this is done for an infinite family of metric spaces further defined. The use of the distance transformation is proposed for it comprises the information concerning all the discs included in a region to obtain fast morphological operators: erosions, dilations, openings and closings (of zero complexity) for an infinite (countable) family of regular metric spaces. New constant time, in contrast with the polynomial time algorithms, for the computation of these basics operators for any structural element are next derived by using this background. Practical examples showing the efficiency of the proposed algorithms, final comments and present research are also given here.
more …
By
Díaz De León S., J.L.; SossaAzuela, J.H.
2 Citations
Mathematical Morphology (MM) is a general method for image processing based on set theory. The two basic morphological operators are dilation and erosion. From these, several non linear filters have been developed usually with polynomial complexity, and this because the two basic operators depend strongly on the definition of the structural element. Most efforts to improve the algorithm's speed for each operator are based on structural element decomposition and/or efficient codification.
A new framework and a theoretical basis toward the construction of fast morphological operators (of zero complexity) for an infinite (countable) family of regular metric spaces are presented in work. The framework is completely defined by the three axioms of metric. The theoretical basis here developed points out properties of some metric spaces and relationships between metric spaces in the same family, just in terms of the properties of the four basic metrics stated in this work. Concepts such as bounds, neighborhoods and contours are also related by the same framework.
The presented results, are general in the sense that they cover the most commonly used metrics such as the chamfer, the city block and the chess board metrics. Generalizations and new results related with distances and distance transforms, which in turn are used to develop the morphologic operations in constant time, in contrast with the polynomial time algorithms are also given.
more …
